Professional Documents
Culture Documents
02 Whole
02 Whole
Year
Ha, Bui Thi Thu, Understanding factors influencing mens readiness to accept IUD for
contraception in rural Vietnam, Doctor of Philosophy thesis, Graduate School of Public
Health, University of Wollongong, 2002. http://ro.uow.edu.au/theses/1892
This paper is posted at Research Online.
from
UNIVERSITY OF WOLLONGONG
by
CERTIFICATION
I, Bui Thi Thu Ha, declare that this thesis, submitted in partial fiilfilment of the
requirements for the award of Doctor of Philosophy, in the Graduate School of
Public Health, University of Wollongong, is wholly my own work unless otherwise
referenced or acknowledged.
TABLE OF CONTENTS
CERTIFICATION
ii
TABLE OF CONTENTS
iii
UST OF TABLES
vii
LIST OF FIGURES
ix
ABBREVIATIONS
ABSTRACT
xi
ACKNOWLEDGEMENT
xiii
1.1. Introduction
10
10
12
13
14
17
19
20
21
22
23
24
26
29
31
33
33
35
38
38
iii
39
41
43
45
55
3.1. Introduction
55
3.2.The setting
3.2.1. Vietnam: country and society
3.2.2.Vietnam: health and health sector
3.2.3. Family planning program in Vietnam
3.2.4. Study population
3.3. Measures
55
56
60
61
64
65
3.4. Implementation
3.4.1. Funding
3.4.2. Ethics approval
3.4.3. Time-line
3.4.4. Recmitment
3.4.5. Consent
3.4.6. Data collection
3.4.7. Training
3.4.8. Field supervision
65
65
66
66
66
66
67
67
68
68
69
70
4.1. Introduction
70
70
71
4.4. Test-retest
71
4.5. Measures
4.5.1. Stages of change algorithm
4.5.2. Decisional balance (pros and cons) measures
4.5.3. Self-efficacy measures
4.6. Statistical analysis
72
72
73
74
74
4.7. Results
4.7.1. Factor analysis of decisional balance scale
75
75
iv
findings
81
84
85
85
86
87
87
88
89
91
5.1. Introduction
91
92
92
92
93
94
94
95
95
96
96
5. 5. Data analysis
99
5.6. Results
102
5. 6.1. Participants' characteristics
103
5. 6. 2. Participants' contraceptive knowledge
109
5.6.3. Communication on family planning issues
112
5.6. 4. Social cognitive factors
114
5.6. 5. Practice of contraceptive methods
115
5.6.6. Factors associated with men's readiness to accept lUD
117
5.6.7. Predictors of stages of men's readiness to accept lUD for contraception
122
5.7. Summary of
findings
129
133
133
137
137
138
138
139
140
6.5. Conclusion
144
145
147
7.3. Measures
147
148
7.5.Resuhs
7.5.1. Test of equivalence
7.5.2. Change in contraceptive knowledge and communication level
7.5.3. Change in stages of men's readiness to accept lUD for contraception
7.5.4. Change in decisional balance and self-efficacy for contraception and
lUDuse
150
150
156
161
163
7.6. Summary of
169
CHAPTER 8: DISCUSSION
8.1. Overview
findings
172
172
VI
LIST OF TABLES
CHAPTER 4
Table 4.1. Staging algorithm for lUD use
73
Table 4.2.Items to assess pros and cons for contraception in general and lUD use. 76
Table 4.3. Retained decisional balance scale items and factor loading
80
81
83
84
85
CHAPTERS
Table 5.1. Cronbach's a coefficient of cognitive scales at baseline study
95
104
106
108
109
110
Ill
Ill
112
113
114
115
116
Table 5. 14. Percentage distribution of men's stage of change for lUD use
116
118
119
121
Vll
125
128
CHAPTER 6
Table 6.1. Percentage distribution of information recalled by participants
141
141
CHAPTER 7
Table 7. 1. Internal consistency reliability, means and SD of scale scores at posttest
148
Table 7. 2. Percentage distribution of socio-demographic characteristics at baseline,
by study group
152
154
156
157
161
163
Table 7.10. Baseline and posttest means on pros and cons for lUD use
167
169
vm
LIST OF FIGURES
Figure 1 . Standardized scores of pros and cons for contraception in general and
lUDuse
86
Figure 2. Standardized self-efficacy scores for contraception in general and lUD use
87
Figure 3. Stages of men's readiness to accept lUD for contraception
117
142
142
143
144
158
158
Figure 10. Change in high communication with wives on family planning issues 159
Figure 11. Change in high communication with others on family planning issues 160
Figure 12. Change in SOC for lUD use
162
164
165
166
168
169
IX
ABBREVIATIONS
AVSC
EngenderHealth International
CPR
DHS
FGDs
FP
Family Planning
GTZ
HBM
ICPD
lUDs
Intrauterine Devices
MAP
Men as Partners
MCH
MOH
Ministry of Health
NCPFP
NGOs
PID
SCT
Social-Cognitive Theory
SOC
Stage of Change
TPB
TRA
TTM
Transtheoretical Model
UN
United Nations
UNDP
UNFPA
UNICEF
ABSTRACT
Background
Research suggests that family planning acceptance is likely to be more effective
when men are actively involved. The goal of the study was to identify methods in
which targeted health behavioural change programs for increasing men's acceptance
of modem contraception might be best developed and implemented.
The
in
the
precontemplation
contemplation/preparation
stages,
stage,
and
10.6
59.5
percent
percent
of
men
in
the
of
men
in
the
xu
ACKNOWLEDGEMENT
To my supervisor. Associate Professor Rohan Jayasuriya and my advisor. Professor
Neville Owen, thank you for your patience, inspiration, insight and intellectual
guidance.
To the Hanoi School of Public Health, especially Dr. Le Vu Anh, the Dean of
School, thank you for giving me the opportunity to do the study, and particular
appreciation to Ms. Le Thi Vui for your invaluable contribution of doing the data
entry for the research.
Special thanks of course to Dr. Peter Caputti for the invaluable statistical advice for
much of my works, to Ms. Julie Marlow for English editing and to Ms. Nancy
Hampel for her comments on the chapter presenting the posttest study.
Many of my work colleagues have been instrumental in providing a positive,
fi-iendly and happy work environment for which I am grateful. For this I would like
to thank Nguyen Due Thanh, Le Thanh Huong, and Pham Viet Cuong at the Hanoi
School of Public Health.
My sincere thanks to my colleagues at the An Hai District Health Centre, Quoc
Tuan and An Hong commune health centres. Without their help, the research could
not have been completed.
Special thanks of course to my parents and my parents- in-law, who looked after my
daughter during the time the study took me away from home.
Finally, I would like to dedicate this thesis to my husband, Ngo Due Dung, and my
daughter, Ngo Ha Anh: to you both my love and gratitude, for without your
persistence, love and support, the time and enthusiasm would never have been
found to finish the study.
XIU
Many studies have examined TTM in different fields. It is argued that the
interventions targeted to a person's stage of change are more likely to be effective
than those that are not (Prochaska et al. 1992). Interventions based on TTM have
been able to combine good efficacy rates and good participation rates and can be
disseminated to whole populations (Vehcer et al. 2000). Studies that have used the
TTM in family planning behaviour have mainly been on condom use, and
respondents' behaviours may have been confounded by their perceptions on risk of
HIV/AIDS (Grimley et al. 1995; Galavotti et al. 1995). In contrast, the lUD is used
only as a contraceptive and is not confounded by the need for protection from
HIV/AIDS.
Vietnam began its first population and family planning program in 1963 and has
achieved a high contraceptive prevalence rate (CPR) of 73 percent (National
Committee for Population and Family Planning (NCPFP) 2001). However, about
one-fourth of couples use unreliable traditional methods such as periodic abstinence
and withdrawal, even though many types of modem contraceptives have become
available.
Men are often significantly involved in contraceptive decision - making and the
degree to which they share in decision-making with their wives have a definite
impact on contraceptive behaviour (Salway 1994; Biddlecom et al. 1996; Bankole
and Singh 1998). In traditional societies like Vietnam, men are the main decision makers in reproductive health matters, including family planning (Johansson et al.
1998b). Regardless of whether the method is one in which the man participates
most actively in its use (such as the condom) or whether the wife participates most
actively in its use (such as lUD), men can play an important role in the method's use
and effectiveness. In spite of the absence of direct targeting of men by the national
family planning program, a vast majority of men approve of family planning (Care
Intemational and Ministry of Health (MOH) 1997; Mai and Montague 1998; Mai et
al. 2001).
The research to date in developing countries, including Vietnam has focused mainly
on men's fertihty preferences and has been descriptive in nature (Ezeh et all996;
Bankole and Singh 1998; Johansson 1998; Mai and Montague 1998). Minimal
2
Chapter 3 describes the research methods and the setting used for the study.
Information on health, health services and the family planning program in Vietnam
is presented.
Chapter 4 describes the pilot study, how the TTM measures were developed and
validated in the context of rural Vietnam, and what changes were made before the
implementation of the baseline study.
Chapter 5 describes the baseline study including research methods and findings on
significant predictors of readiness at each stage for men to accept the lUD for
contraception. These predictors served as the basis for designing the stage-targeted
intervention for promoting men's acceptance of contraception in general and lUDs
in particular.
Chapter 6 explains how the stage-targeted intervention program for promoting
men's readiness to accept lUDs for contraception was developed and carried out,
and presents the results of evaluation of the stage-targeted letters
Chapter 7 presents the impact of the stage-targeted intervention for promoting
men's readiness to accept lUD for contraception after six months follow-up.
Chapter 8 presents the discussion of the findings of the intervention study. It
contains a summary of research findings, a discussion of features and limitations,
possible alternative explanations for the findings and implications for family
planning program in Vietnam and further research.
Traditionally, men are powerftil in all areas of reproductive health despite the bias
of family planning services towards women (Green et al. 1995; Drennan 1998). The
majority of men consider themselves as the primary decision-makers regarding
confraception (Green et al. 1995; Johansson et al. 1998b). In male dominant
cultures, where women are not permitted to leave the home without male company,
the opinion of the husband has greater weight in the decision-making process, in
obtaining suppUes and method to be used (Wells 1997). A study of fertility
decisions of five generations of one South Indian family found that men tend to
confrol confraceptive use and to make fertiUty decisions (Karra et al. 1997). The
men in the older generations chose to limit their own fertility by having vasectomies
without telling their wives.
Ezeh (1993) reported the same pattern in Ghana, where husbands exercised
exclusive confrol over family planning. Men's confraceptive attitudes operated
through their mate selection and cultural norms that subjugate women to men.
Moreover, some men required their wives to obtain their consent before seeking any
confraceptive method.
Among men who supported the confraception in Nigeria, a majority beheved that
responsibility for confraceptive use rests with women (Oni and McCarthy 1991;
Obionu 1998). A similar situation was reported in Vietnam (Anh et al. 2002).
Ideally, a couple's decision about family size and confraceptive use should be made
jointly, with equal consideration given to the concerns of the husband and wife.
However, even within the same culture, different opinions exist with regard to
responsibility ranging from total male domination to complete male indifference.
Mason and Smith (2000) analysed data from five countries namely the Philippines,
Thailand, Pakistan, Malaysia and India, and found that men in highly gendersfratified settings often confrol whether or not wives use confraception. In such
communities, the husbands' preference on the use of confraceptive is more likely to
prevail.
Men's role is dominant in sfrongly patriarchal societies, but tends to diminish or
disappear in more egalitarian environments (Necchi 1999). In more egalitarian
7
reproductive health choices, to safeguard thefr partners' and thefr own reproductive
health and to enhance couples' access to male confraceptive methods (Wegner et al.
1998).
The term 'involvement' connotes participation or engagement. Verme et al. (1997)
posits term implies that men are uninvolved. Nevertheless, they are, arguably,
already too involved in reproductive health as policy makers, service providers, or
husbands. Men's involvement refers to any activity that seeks to expand the
provision of reproductive health services and information, to include males of all
ages, either individually or as part of sexually active couples (Danforth and Green
1997). Whatever the term used, the purpose is to describe a complex process of
social and behavioural change that is needed for men to play more responsible role
in reproductive health/family planning. Men's participation/involvement can be
seen as a means to an end, rather than as a goal in itself (Drennan 1998).
Green (1994) defines several ways to get men involved in family planning services
such as providing alternatives for couples dissatisfied with their current methods by
increasing male confraceptive use; promoting greater discussion between partners;
and changing male attitudes regarding confraception, thereby enabling women to
practice confraception.
1.5. Key individual factors influencing contraceptive behaviour
This literature review has identified several socio- demographic factors as well as
the sorts of knowledge, attitudes and practices that influence men's confraceptive
behaviour or family planning acceptance.
1.5.1. Men's knowledge of contraceptive method
Men's knowledge of confraceptive methods is high, but it varies by countries and
regions. This finding is consistently reported in all Demographic Health Surveys
(DHS) of Afiica and Asia (Ezeh et all996; Drennan 1998; Bankole and Singh
1998). In almost all countries, surveyed men are more likely than women to know
about confraception. In East Afiica, North Africa and Asia, more than 90 percent of
10
men know at least one confraceptive method, with the exception of Pakistan and
Tanzania where only 79 and 86 percent of men, respectively, know a method.
Knowledge is lowest in West Afiica. In Bangladesh, Brazil, Haiti and Pakistan, the
knowledge level is quite similar among men and women (Ezeh et al. 1996; Ezeh
and Mboup 1997; Drennan 1998; Bankole and Smgh 1998).
The demographic health surveys, however, report only whether respondents have
heard of the various confraceptive methods (obtaining spontaneous or prompted
answers).
respondents know how to use a method correctiy (Ezeh et al. 1996). Bongaarts and
Bmce (1995) propose a more comprehensive approach to measure knowledge of
confraception. A person is considered to have an acceptable knowledge of a method
if he/she can describe how it is used, its main side effects and where it can be
obtained.
Information about the relationship between use of modem confraception and
knowledge on confraception is available (National Research Council 1993).
Analysis of the World Fertility Survey and DHS data for selected countries in subSaharan Africa indicates that where knowledge levels for the region are less than
80.0 percent, use of modem confraceptives is less than 10.0 percent among couples
in union.
In Vietnam, the DHS surveys have not collected information from married men.
Therefore, it is quite difficult to compare knowledge levels in Vietnam with other
countries.
knowledge levels for men (Care Intemational and MOH 1997; Mai and Montague
1998; Mai et al. 2001).
In most countries, men are more likely to know of modem confraceptive methods
than fraditional methods (Ezeh et al. 1996; Ezeh and Mboup 1997; Mai and
Montague 1998; Drennan 1998).
confraceptive pill (the pill), followed by condoms and female sterilization. Of the
fraditional methods, periodic abstinence is better known than v^ithdrawal. Men are
more likely to know about female sterilization than vasectomy. In Bangladesh,
11
13
More than half of all currently married men in Bangladesh, Egypt and Kenya
currently use a confraceptive method (Ezeh et al. 1996; Drennan 1998). The lowest
level is reported in West Afiican countries. In most countries, the pill and condom
are the most widely used methods, often accounting for more than 90 percent of all
modem method use.
15
couples rely on condoms, and lowest in Sub-Saharan Afiica and North Africa
(Drennan 1998).
In Bangladesh, despite increasing awareness, the use of condoms declined from
14.6 to 8.2 percent during the period 1975-1993. The main reasons related to health
concerns, ineffectiveness, inconvenience, and reduced sexual pleasure (Donahoe
1996). Similar reasons were found by Jahan et al. (1996) in an exploratory study in
an urban area of Dhaka. In addition to the above-mentioned reasons, problems with
availability and their storage/disposal were also barriers against condom use in India
(Khan and Patel 1997).
In Vietnam, condoms are mainly distributed through the social marketing system
and private sector. Non-profit government organizations (NGOs) play Uttle role in
condom distribution. Despite the widespread social marketing program, condom use
is still limited. In the last ten years, the rise of condom use was very modest from
1.2 percent in 1988 to 6.0 percent in 1997. The main reasons for low usage are
similar to those in Bangladesh and India (Care Intemational and MOH 1997; Mai
and Montague 1998; Huy 2000). According to the United Nations Population Fund
(UNFPA) (2000), one important reason relates to its high cost compared to average
income, particularly for rural people.
Periodic abstinence (rhythm) and withdrawal
Periodic abstinence and withdrawal are the oldest methods for prevention of
pregnancy. Periodic abstinence involves avoiding sexual relations during the fertile
period of the menstmal cycle. Withdrawal is coitus interruptus. The decision to
use these methods needs to be made jointly by couples, and it requires awareness of
basic conception/fertihty physiology, including menstmal cycle (Gallen and Liskin
1986; Douthwaite 1998). The use of these methods is often the outcome of a sfrong
desire to regulate fertility but they have a high failure rate. Some studies found the
high failure rate was due to lack of knowledge of the menstmal cycle (Gallen and
Liskin 1986) or lack of male self-confrol (Douthwaite 1998).
16
The two fraditional methods are not widely used in developing countries except
Vietnam, Turkey and South Korea. Recently about 3 percent of couples in
developing countries were shown to rely on withdrawal and 4 percent on rhythm for
prevention of pregnancy (Drennan 1998). However, percentages vary largely
among countries. Worldwide, rhythm is the least used of methods involving male
cooperation. However, DHS surveys might underestimate the use of fraditional
methods, because they do not report the combined use of fraditional with modem
methods such as the pill or condoms.
In Vietnam, the practice of periodic abstinence and withdrawal is substantial. It
tended to increase during the period from 1988 to 1997 (NCPFP 1999). The sttidy
on confraceptive mix by Nhan et al. (1999) found a large number of couples in
Vietnam switched from modem methods to the fraditional methods due to health
concerns. By 1997, around one-sixth of ever-married women said that they had
practiced rhythm at some time and one-fourth said that they had practiced
withdrawal. Current use of these methods is also high. About 7 percent of
currently married women report that they practice periodic abstinence, and 12
percent use withdrawal.
The practice of withdrawal is more common than periodic abstinence. The
combined prevalence of these two methods accounts for ahnost 25 percent of
overall current confraception. This is one of the highest proportion shares in any
developing countty with moderate to high CPR (Phai et al. 1996; NCPFP 1999).
Interestingly, urban and highly educated people are more likely to use fraditional
methods than rural people (NCPFP 1999). Given the high usage rate despite its high
failure rate, Johansson (1998) emphasizes the remarkable lack of attention given to
men by family planning programs.
1.5.5. Use of lUDs in developing countries and Vietnam
Globally, the lUD is the most common reversible confraceptive method used by
women. It has been used for more than three decades (Hicks 1998). The first
modem lUDs appeared in the early 1960s and since then, different generations have
been produced. Nowadays, the most widely available lUD is the Tcu-380. The
17
United States Food and Dmg Adminisfration approved this model in 1994 for use
up to 10 years. This is recognized as the longest-lasting copper lUD. The Tcu-380 is
one of the most effective methods of confraception ever developed. Fewer than one
woman per 100 become pregnant in the first year of use and only 2.1 per 100
became pregnant in 10 years of use (Treiman et al. 1995). Tmssell et al. (1995)
estimated the costs of 15 different confraceptives and concluded that the lUD is the
most cost-effective reversible method. Despite its high initial costs, the lUD
becomes more cost-effective the longer it is used.
The highest prevalence of lUD use is reported in China and Vietnam, where 30
percent of married women use lUDs. In Vietnam, the lUD accounts for almost twothfrds of all current modem contraceptive use (NCPFP 1999). lUDs are also widely
used in other Asian countries, particularly in Indonesia and in Taiwan (Treiman et
al. 1995).
There are some problems associated with lUD use, e.g., expulsion and infection.
Expulsions usually occur in the first year, especially in the first 3 months after
insertion, however, the expulsion rate is relatively low at 1.6 to 8.0 per 100 users
(Hicks 1998). The risk of infection or pelvic inflammatory disease (PID) is its
greatest risk and is associated with the preponderance of anaerobic organisms. The
risk is greater for women with multiple sexual partners than those with a single
partner. For parous women, the risk is from bacteria infroduced at insertion. The
relative risk is 6 times higher in the first 20 days. Moreover, the use of lUDs has
not been associated with increasingfransmissionor acquisition of HIV/AIDS, nor is
confraceptive efficacy reduced among HIV infected women.
Analysis of confraceptive discontinuation based on DHS from six countries
(Morocco, Tunisia, Egypt, Ecuador, Indonesia, and Thailand) showed the
continuation rate of lUD use was 82-89 percent after one year and 65-80 percent
after two years (Ali and Cleland 1995). The health concerns including side effects
related to lUD use are much lower than those of other modem methods. Hieu et al.
(1995) reported a similar continuation rate of 81.0 percent of lUD use after one year
in Vietnam. The main reasons of termination were expulsion (9 percent) and health
problems (7 percent). However, Johansson et al. (1998a) found a much higher
18
expulsion rate of 14.0 percent; 30.0 percent and 44.0 percent after first, second and
third year, respectively. The deliberate removal of lUDs was the main reason for
the high expulsion rate among users in Vietnam. This was done under the guise of
method failure in the hope of satisfying the family's wish for a son while avoiding
the criticism of exceeding the two-child limit.
In Vietnam, the government has promoted lUD use since the 1960s using the target
(quota)-system (Noble 1996). In the last few years, the government has tried to
increase the confraceptive mix. However, in the rural areas, few other confraceptive
options are available for people who want to prevent pregnancies (Toan et al. 1996;
Noble 1996; Johansson et al. 1998a; Thang and Huong 1998).
1.5.6. Abortion
About 32 million induced abortions per year occur in developing countries
(Bongaarts 1997). Asia accounted for 24 millions, Latin America account for 5
million and Afiica account for 4 million. Most abortions are illegal in Afiica and
Latin America, while about one-third is illegal in Asia. About one in four
pregnancies are ended by abortion in developing countries.
The abortion rate has a correlation with CPR. In most urban regions of Mexico and
Colombia the abortion rate declines as confraceptive use stabilized or increased
(Singh and Sedgh 1997). The studies show that confraception is the most effective
way to reduce the likeUhood of an unwanted pregnancy, birth and abortion.
Confraception reduces the probability of having an abortion by 85 percent (Cohen
2000). In two countries sharing the same birth rate, the one with the higher CPR
will see fewer abortions. However, the rise in abortion incidence may continue until
access to confraceptive choice improves (Kulszycki et al. 1996).
Abortion has been legal in Vietnam since the 1960s and currently it is widely
available (Belenger and Hong 1998). Vietnam is among those countries with the
highest abortion rate: 2.5 abortions per woman hfetime (Goodkind 1994). The main
reason relates to increasing premarital sex among yoimg people. However, a
19
21
The DHS data from 18 countries showed that between 1990 and 1996 married men
on average wanted a larger number of children in many of these countries.
Husbands in all surveyed countries except Pakistan wanted fewer than five children.
The average preferred number of children desfred by wives showed a similar range
across countries and regions (Bankole and Singh 1998). The number of children
desired by men appeared relatively small in Bangladesh: 2.5 children. It was higher
in Pakistan at 4.1 children (Ezeh et al. 1996). By the ages 30 to 34 years, nearly
half the men did not want any more children.
In Vietnam, the results from DHS surveys and other studies show a substantial
lowering of desired number of children among women (Phai et al. 1996; Thang and
Huong 1998; NCPFP 1999). hi the past two decades, the number declmed from 5.5
to 2.8 for currently married women. The same number was reported by men (Mai
and Montague 1998). The small desired number of children is consistent with
fertility decline in Vietnam and this could be a result of population poUcy on 'one or
two child' (Gammeltoft 1999). Although some more recent qualitative studies
report the prevalence of the desire for two children that is the norm for a majority of
couples (Anh et al. 2002), many people could argue that the report for two children
could be partially a result of respondents' providing social desirable answers, or of
their fear of a government penalty (Phai et al. 1996; Gammeltoft 1999).
1.5.9. Fertility intention
Another prominent measure of reproductive preference is whether or not the
respondent intended to have another child. This measure is a robust predictor of
confraceptive and fertihty behaviour at both the aggregate and the couple's levels,
and it is an important measure of unmet need for confraception (Westoff 1990).
However, Bongaarts (1992) states that under-reporting of unwanted births is
common due to women's reluctance to classify their offspring as unwanted.
Confraceptive use is more prevalent among people who want to limit childbearing
than those who only want to space children (Bongaarts 1992). In societies where
relatively few women want to limit childbearing, fertility intention has only a small
impact on confraceptive use and fertility. This pattern is more prevalent in Sub22
Saharan Afiica, where the data show that a high proportion of husbands want both a
large family and the next child sooner than their wives'. In comparison, in countries,
where a large proportion of married women want to stop childbearing, most of them
practice confraception. The use of confraceptive methods is highest when couples
agree to stop childbearing; what happens when there is disagreement is unclear.
The effect of the husband's desire for another child is inconsistent across studies.
While Salway (1994) found that the husband's desire for no more children has no
effect on contraceptive use, Speizer (1999) reported that its has a significant role
even after confroUing for women's fertility desfres.
1.5.10. Son preference
Son preference has an influential impact on fertility and confraceptive behaviour
and it has been consistently reported in many studies across countries (Oyeka 1989;
Arnold 1992; Rajaretnam and Deshpande 1994; Haughton and Haughton 1995;
Wongboonsin and Ruffolo 1995; Stash 1996; Graham et al. 1998; Hussain et al.
2000; Nosaka 2000; Yount et al. 2000). Sons are important because they can inherit
wealth, provide security for parents in older age, and continue the family name
(Mwageni et al. 1998).
The DHS data in 26 countries show that parental preference for sons persist in some
countries but may be decUning in others. The most common preference pattern is
for at least one son and one daughter (Arnold 1992). Son preference remains sfrong
in South Asian countries, particularly those influenced by Confucianism
(Wongboonsin and Ruffolo 1995). However, rather than an exclusive son
preference, couples strive for one or more sons and at least one surviving daughter
(Hussain et al. 2000). Confraceptive use is less likely among couples that want to
continue trying for the desfred number of sons. In countries where both
confraceptive use and smaller family sizes are increasingly popular, son preference
has the potential to bolster higher rates of fertility and lower rates of confraceptive
use (Stash 1996). In Nigeria, women with no living sons are least likely to use
modem confraceptive method and its use increases dfrectly with the number of
hving sons (Oyeka 1989).
23
Son preference can be seen as undermining the success of the overall development
process, because it reflects discrimination on the basis of sexfromthe earhest to the
later stages of hfe (UN 1995). In general it may lead to problems of sex
discrimination, sex-selective abortion, female infanticide, a poor quality of life for
females, a marriage squeeze, deterioration of the family system and have effects on
future fertility (Wongboonsin and Ruffolo 1995). Furthermore, Edlund (1999)
states that increasing occurrence of sex determination in societies favouring sons
may lead to social segregation of the sex, with men at the top and women at the
bottom. This causes a surplus of males, a situation that has arisen in parts of China.
Social and cultural factors are likely to be substantial determinants of son preference
in South East Asia (Wongboonsin and Ruffolo 1995). According to the world
standard, Vietnam has a sfrong son preference and it results from a blending of
Conflician and bilateral kinship systems, socio-economic factors, population policy
and politicalfransformationon gender relations. However, it has a minor effect on
fertility due to high abortion rate in Vietnam, The high abortion rate may reduce the
role of CPR in reflecting son preference (Haughton and Haughton 1995). If
abortions were absent, fertility would fall by roughly 10 percent from the current
level of 3.2 children per woman.
Due to the 'one or two-child policy', couples want to make sure that they bear some
sons (Goodkind 1994). Couples with daughters are 27 percent less likely to use
confraception than those who have children of both sexes. Couples with three and
more children are more Hkely to use modem methods than those with fewer
children (Anh 1995). However, with rapid socio-economic changes, son preference
tends to be declining and disappearing in Vietnam (Ha and Schuler 1999; Haughton
2000)
1.5.11. Men's unmet need for contraception
Unmet need for confraception is a global issue. This has been one of the most
widely discussed family planning concepts in recent years. Millions of people in
developing countries who would prefer to postpone or avoid pregnancy do not use
24
confraceptives (Bryant et al. 1996). The first measures of unmet need developed by
Westoff has become standard (Bhushan 1997). Recently the methodology for
measuring unmet need for confraception has become increasingly refined and more
complex (Bongaarts and Bmce 1995).
Unmet need is based on women's responses to survey questions. When women
respond that they want to postpone or avoid childbearing but are not using
confraception, they are defined as having an unmet need. The standard formulation
of unmet need has been applied more exclusively to married women. However,
there have been criticisms of this formulation. Mueller and Germain (1992) suggest
that the unmet need concept should be apphed to all sexually active people,
regardless of marriage status.
focusing on couples' unmet need has a greater impact on fertihty than focusing
solely on women. Becker (1999) infroduced the concept of couples' unmet need,
defined as the proportion of couples with at least one partner having an unmet need
for confraception. A minimum estimate of unmet need for couples is produced when
both partners have unmet need; a maximum unmet need occurs if the couple unmet
need rests with one spouse only. The results showed that in the Dominican
Republic, Bangladesh and Zambia, unmet need for married women was
significantly different from couples' and from men's only (Becker 1999). The
discrepancies may have indicated spousal disagreement or lack of communication
about reproductive goals or contraceptive use. Therefore, in order to reduce unmet
need among couples, particularly in male dominant cultures, family planning
programs needs to target men, encouraging small families, while impUcitly
acknowledging men's power in confraceptive decision-making.
Ross and San (1997) analysed 1988 DHS data and found that unmet need was
substantial among couples in Vietnam. Unmet need existed among one-sixth to over
one-fourth of couples, without counting categories such asfraditionalmethod users,
dissatisfied users, abortion cases, or the unmarried. Couples with an unmet need
were more hkely to be yoimg, living in rural areas, of low education, and with small
famihes without sons. Despite the substantial use of abortion, a very high
proportion of couples had more children than they desired.
1.6. Couple communication and decision-making on contraception
Couple communication on family planning matters is a significant predictor of
confraceptive use and it allows shared decision-making and more equitable gender
roles (Drennan 1998). The majority of studies in many countries support the
finding. In Pakistan, Nigeria, and Ghana, couple communication has been correlated
with current confraceptive behaviour (Oni and McCarthy 1991; Mahmood and
Ringheim 1997; Odimegwu 1999).
its impact on the exposure to the risk of pregnancy, desfred family size and
confraceptive practice (WilUams et al. 1999).
Similarly, a positive relationship between income and confraceptive use is reported
in a majority of studies (Oni and McCarthy 1991; Nustas 1999; WilUams et al.
1999). In Jordan, men who come from families with high incomes were more likely
to believe that they should practice confraception than those with low incomes. In
Nigeria, men's support for family planning increases from 46 to 78 percent with
increasing income levels. One-third of men in the poorest areas, with no formal
education reported willingness to use confraception, while two-thfrds in areas with
higher incomes and higher education reported their willingness.
Some studies report the positive influence of occupation on confraceptive use. For
example, Takyi and Sakyi (1997) found that in Ghana professional husbands were
more likely to have lower fertility and more likely to practice confraception than
agricultural workers. In Bangladesh, confraceptive use was likely to be 1.8 times
higher among couples whose husbands were employed in sales, services or
production than those who were agricultural workers (Islam and Mahmud 1995). A
similar frend is observed in Vietnam, where employed people are more likely to use
confraception than those who are unemployed (Haughton and Haughton 1995).
Geographic difference also plays a significant role in confraceptive acceptance.
Rural men are more likely to desire more children and less likely to approve
confraception than urban (Ezeh et al. 1996). However, not all studies support the
frend. For example, in Tanzania, Mwageni et al. (1998) did not find a significant
difference in attitudes toward confraception among mral and urban men. This may
be due to similar cultural norms in the two regions.
ReUgion is often citied as a factor that significantly influences confraceptive
behaviour. Certain religions such as Catholicism are more disapproving of
confraception than others like Judaism and Protestantism. In countries where Islam
is sfrong, men are less likely to use confraceptive than other countries (Drennan
1998). Douthwaite (1998) reported that in Pakistan, Islam often is quoted as a
barrier to confraception. A similar frend is observed in Sierra Leone, where Islamic
30
shows that many elements of quality of care are ignored (Nhan et al. 2000). The
counselling, and technical competence of health providers is generally poor.
Recently, the government of Vietiiam started emphasizing the quaUty of care in
family planning services and this has been reflected in the country's reproductive
health sfrategy 2001-2010. However, it is challenging and many elements need to be
improved.
1. 9. Strategies to increase male involvement
Different programs and projects that involve men in family planning have been
carried out with considerable success. The main goal of most programs is to
improve reproductive health by encouraging responsible sexual behaviour, and use
of male confraceptive methods; to create greater male support for partners' actions;
to improve couples' communication; and to provide education for youth (Green
1994). Several sfrategies could be designed to get men involved with low cost and
ease of implementation.
1.9.1. Designing service
Men's interests and needs regarding confraceptive use differ from women's in some
aspects. To meet the needs of men, providers should offer an array of services, treat
men with sensitive counselling, respect their privacy and confidentiality, have easy
accessibility, and be low in cost (Green et al. 1995). However, there are no models
for men's reproductive health services comparable to the existing well-defined
constellation of obstetric and gynaecologic services for women (Wegner et al.
1998). The AVSC Intemational recommended a model developed in the USA. The
model consists of three categories of services: (1) screening which includes the
information to be obtained from each male using the clinic, as well as information
about what services he may need; (2) information, education and counselling
services; and (3) diagnostic andfreatmentservices. However, this model needs to be
revised and adapted to specific existing health delivery systems in developing
countries.
33
The other way to increase the availability of services is to adapt the existing services
to make them more accommodating and attractive to men. Possible intervention
includes dedicating hours for men, inviting men to come with their partners, hiring
male health workers and educators. In Pern, male distributors reach more men and
sell more condoms than their female counterparts. A similar sfrategy was
implemented by USAIDS in Kenya, MaU and Honduras (Green 1994; Danforth and
Green 1997). Furthermore, the scarce resources in most developing countries
suggest integration of services is more cost-efficient (Wegner et al. 1998).
Mobile services
A mobile service is an oufreach service that is attached to an existing clinic. It is a
very useful approach for reaching men in the community. Experience shows that
mobile services are not only important in rural areas where there are no existing
facilities but also for urban neighbourhoods (Wegner et al. 1998). It affords an
opportunity to reach men directly and individually with convenient services and
accurate information. This approach has been successfiilly implemented in Pakistan
through the project 'Reaching men in the community' and, in MaU through the
Katibuogou project (Green 1994).
Workplace programs
Reproductive health services also need to be designed to fit workers' needs by
rearranging opening hours (Wegner et al. 1998). The success of Family Health
34
Intemational in Kenya suggests that the use of peer educators to reach men in the
workplace is a very useftil approach. In confrast, Marie Slopes Intemational
includes information on reproductive health in the curriculum of the National Youth
Service and this helps to reach thousands of young men in Kenya.
1.9.2. Communication
Men are often not aware of their own reproductive health needs nor of the
availability of services; therefore, it is cmcial to use a variety of communication
sfrategies to help men gain access to both information and services (Wegner et al.
1998).
Information programs can help to promote discussions between partners and lead to
healthier family planning practices. To be appropriate for male audiences, messages
should promote behaviour change and offer information and services that men want
(Drennan 1998).
In the family planning area, communication can be divided into three broad
channels: interpersonal, including family, fiiends, and health care providers; group,
including mobilisation of community organisations; and mass media, including
print and broadcast.
At the interpersonal level, various studies have shovm that providing men with
information and counselling helps them be more supportive of confraceptive use,
and more aware of the concept of shared decision-making (Wells 1997). Terefe and
Larson (1993) stated that involving the husband in family planning education in the
home significantly increases use of modem confraceptives
in Ethiopia.
Confraceptive use after a one-year intervention nearly doubled among couples who
received husband-wife counselling, compared with use among couples in which
women were counselled alone.
women whose husbands did not receive counselling (Amatya et al. 1994). Ozgue et
al. (2000) show clear evidences of increasing confraceptive use in Turkey when the
educational intervention involves both men and women.
Lessons leamt from communication intervention suggest that it is cmcial to
ascertain the needs of men and determine what communication approaches are
acceptable or appropriate. Messages must address the misconceptions of both men
and women and need to be gender-sensitive (Piofrow et al. 1997; Wegner et al.
1998). A study in Kenya suggested that health providers should anticipate men's
outspokenness and understand the male agenda if they were to counsel men
effectively (Kim et al. 2000). The counsellor should be frained to have knowledge
about masculinity, gender roles and values, about men's position in the family and
to talk openly about sexual behaviour (Wegner et al. 1998).
At the community level, the studies show that community involvement is critical to
a program's success (Wegner et al. 1998). Some community groups can play a
major role in conducting service delivery and education programs. Therefore, the
employment of these groups could help to improve the family planning services. In
Cameroon, group of leaders have been employed to make home visits and give
educational talks. This approach contributes to a significant increase of knowledge,
approval and use of contraceptive among couples in the village (Green 1994). This
similar sfrategy has been implemented by USAIDS in some countries including
Egypt, Bangladesh, Nigeria and Oman (Danforth and Green 1997).
A multimedia approach using television, radio, newspapers, and magazines has
been found to work best in family planning communication (Piofrow et al. 1997)
because it communicates to a wide audience (Wegner et al. 1998). In Kenya, the
multimedia approach includes radio and television spots, newspaper advertisement,
leaflet, booklet and poster distributions, this campaign increased the vasectomy rate
by 125 percent.
Mass media are increasingly cost-effective and practical for reaching large numbers.
In Zimbabwe, a communication campaign included television, radio, newspapers,
magazines, football matches and community mobiUzation events and reached over
36
half million men; increase in use of modem confraceptive methods was found to be
up to 20 percent greater than the previous frend (Piofrow et al. 1992).
Radio and television are probably more cost-effective in reaching populations than
field workers. In addition, the mass media can depict scenes and bring up subjects
that field workers may be unable or reluctant to discuss with people face to face.
However, since resources are always limited, communication projects should
identify a leading medium to carry the message and focus major efforts. This
decision should be based primarily on the initial audience analysis (Piofrow et al.
1997). Although mass media can raise awareness, they do not necessarily lead to
behaviour change. There is a need for reinforcement through other means of
communication such as counselling and written material. Communication sfrategies
should address the diverse needs of different groups (Wegner et al. 1998).
In summary, the key individual factors influencing confraceptive behaviour, such as
knowledge, approval, practice of confraception, types of sexual relationship, fertility
intention and son preference, were reviewed. The roles of socio-demographic
factors, e.g. education, occupation and income, were reported in relation to
confraceptive
37
38
illness health behaviour and sick-role health behaviour (Conner and Norman 1995).
However, some health behaviours do not fall within these categories such as
medical service usage (vaccination, screening), compliance with medical regimens
(dietary, diabetic) and self-directed health behaviour (diet, exercise). Confraceptive
behaviour is classified as self-dfrected health behaviour for self-management or
improvement of well-being.
Four most commonly used behaviour change models that apply to confraceptive
behaviour at the individual level are: the Health BeUef Model (HBM), the Theory of
Reasoned Action (TRA), the Theory of Planned Behaviour (TPB) and
Transtheoretical Model (TTM) (Glanz et al. 1997). Among these theories, the TTM
is one of the leading models of health behaviour change. The model includes three
theoretical constmcts cenfral to change: (1) stage of change (readiness to take
action); (2) decisional balance (pros and cons associated with behaviour's
consequences; and (3) self-efficacy (confidence to make and sustain change in
difficult situation).
Some social cognitive factors identified by the above theories are important in
understanding family planning behaviour: (1) perceived susceptibility to and
severity of consequences of pregnancy and abortion; (2) perceived benefits and
perceived costs of confraception; (3) attitudes, norms and behavioural intention and
(4) self-efficacy.
2.2. Perceived susceptibility to and severity of consequences of pregnancy
Perceived susceptibility to pregnancy and severity of consequences of pregnancy
was originally derived from the Health BeUef Model (Rosenstock 1974). The model
suggests that the individual weighs the potential benefits of the recommended action
against the psychological, physical, and financial costs (the barriers) when deciding
to act. Individuals also evaluate whether or not they are susceptible to a threat and
whether or not the threat is tmly severe.
susceptibility, and benefits, and the weaker the perception of barriers, the greater the
likelihood the healtii-protective actions will be taken. Overall, perceived barriers are
the sfrongest predictor of whether or not an individual will engage in healthprotective behaviour, followed by susceptibility, benefits and severity (Conner and
Norman 1995).
Perceived susceptibility to pregnancy is a subjective perception of the risk of
becoming pregnant. The perceived susceptibility to pregnancy has been found to be
an important factor in predicting confraceptive behaviour (Hester and Macrina
1985, Eisen et al. 1985). Hester and Macrina (1985) studied confraceptive behaviour
among 213 coUege women and revealed that negative attitudes towards pregnancy
due to its serious and dismptive consequences were common to all women.
However, in that study only perceived susceptibiUty to pregnancy was found to be
an important predictor of confraceptive behaviour.
The role of perceived serious consequences of pregnancy has been inconsistently
reported as a significant predictor for confraceptive behaviour. Eisen et al. (1985)
reported that it was the most important predictor, followed by student's ethnicity.
However, Hester and Macrina (1985) stated that despite recognition of the
seriousness of pregnancy, perceived severity was not an indication that people wiU
adopt contraception.
HBM has been used for an educational intervention program for prevention of
pregnancy among teenagers based on the hypothesis that adolescents who have a
higher perceived threat will be more Ukely to participate in this intervention
program (Eisen et al. 1992). However, the finding did not support the hypothesis.
About 57 percent of adolescents who had been pregnant did not participate in the
program. This confirms the finding in Hester and Macrina (1985) that the
educational program emphasizing adverse consequences of a pregnancy is not likely
to have a substantial impact on confraceptive behaviour.
Perceived risk (severity) of undergoing an induced abortion refers to feeling about
the seriousness of having an abortion. It includes the feeling about the severity of
consequences of an abortion like pain, medical complications and adverse
40
psychological impacts associated with cultural values. This constmct has been
reported as an important factor in predicting confraceptive behaviour (Lucker 1975).
Although perceived susceptibility and severity are reported as significant factors in
predicting health behaviour using HBM, many concems have been expressed
regarding this model. The vast majority of studies using this model have failed to
estabUsh validity or reliability of measurement prior to testing and confusion
remains about relationships among HBM components (Conner and Norman 1995).
HBM also does not take into account other aspects of people's lives such as
significant psychological factors. Moreover, the operationalisations of constmcts
are implemented differently by investigators. Conner and Norman (1995) argued
that the HBM model failed to address some cognitive variables shown to be
important predictors of behaviour such as others' approval of behaviour.
Although the perceived susceptibility to pregnancy and severity of the consequences
of pregnancy/abortion were important predictors in contraceptive behaviour, the
difficulty with the measurement of these constructs limits thefr application in
research. With a more careful constmction, the two factors will be very important
and useful in future research on family planning behaviour (Conner and Norman
1995).
2.3. Perceived beneflts and costs of contraceptive behaviour
Perceived benefits and costs as cognitive factor in behavioural change, has been
derived from different theories and constmcted differently. It appears as 'perceived
benefit' and 'perceived barrier' in HBM and the decisional balance in TTM.
HBM suggests that the individual weighs the potential benefit of the recommended
action against the psychological, physical and financial costs (barriers) of the action
when deciding to act. The cost of confraceptive use encompasses much more than
physical access to family planning services. It includes social, psychological,
medical and cultural barriers to confraceptive practice (Bhushan 1997). To date
relatively little empirical research has attempted to measure all concepts because of
difficulties in making explicit the full array of costs (Casterline et al. 1997).
41
Assessing these barriers is particularly chaUenging if tiie data available are mamly
from cross-sectional social surveys (Conner and Norman 1995). In a cross sectional
study it is not possible to find out whether the action taken to not take a
confraceptive/or discontinue a confraceptive was before or after the 'barrier' wiU
not be known. Recall bias also be a problem when cross sectional surveys are used.
hi HBM studies, perceived barriers have been found more consistently as the
sfrongest predictor of behaviour (Conner and Norman 1995). Among 632 women,
people who perceived less serious side effect were more likely to use the piU
(CondelU 1986). A similar finding was reported by Lowe and Radius (1987) and
Keith et al. (1991). Less perceived barriers was most sfrongly associated with
effective confraceptive behaviour. Confraceptive non-users tend to report more
barriers and less benefits of confraceptive use than users. However, the perceived
costs and benefits were not found as a significant predictor for condom use in one
study among college students (Thompson et al. 1996).
In TTM, the perceived benefits/costs of confraceptives is conceptuaUsed as the
'decisional balance'. This concept has been proposed by Janis and Mann (1977) as
a schema for representing both the cognitive and motivational aspects of human
decision-making. This suggests that the anticipated gains (benefits) and the
anticipated losses (costs) can be categorised into four types: 1) gain or loss for self;
2) gain or loss for significant other; 3) approval or disapproval from significant
other; and 4) self-approval or disapproval.
In TTM, instead of eight factors that need to be balanced when making a decision,
only two factors are considered, namely, pros and cons of a behaviour. The balance
of pros and cons depends on the stages that people have reached. In the early
stages, people judge the pros of the behaviour to outweigh the cons. In the later
stages, the opposite pattem occurs (Prochaska et al. 1994). Grimley et al. (1995)
studied different reasons for confraceptive use among 248 heterosexually active
college men and women and indicated that the highest pros for both general
confraceptive use and for condom use with main partners was protection from
pregnancy, while the highest pros with casual partners was prevention of diseases.
42
The highest cons for using confraception with main partners was the feeling of
having unnatural sex and for condom use, relying on partners' cooperation.
Galavotti et al (1995) designed a sunilar study for women with high risk of HIV and
unintended pregnancy. In this study, different confraceptives other than condoms
were studied. The general confraceptive measure appeared a significant predictor
for use of other specific methods like the pill and Norplant. Therefore, it was
beUeved that it was not necessary to measure specific confraceptive methods. A
similar finding was reported by Hester and Macrina (1985), who suggested that
beUefs and perception about specific contraceptive methods were not critical in
explaining confraceptive behaviour. The beUefs and perception about specific
methods often derived from an information base, while perceptions about
contraception in general were emotionally derived.
2.4. Attitudes, subjective norms and intention in contraceptive behaviour
For Fishbein (1972), a given behaviour is a function of the intention to perform the
behaviour, and an intention in turn, is a joint fimction of (1) the attitude towards
performing the behaviour and (2) subjective norms that govem the behaviour. Those
are core constmcts of the Theory of Reasoned Action (TRA). The concept of
perceived behavioural confrol was incorporated into these three constmcts in the
Theory of Planned Behaviour (TPB) (Conner and Norman 1995). This concept
encapsulates a person's expectancy that performance of the behaviour is within his
or her confrol. The concept is similar to the concept of self-efficacy. Individuals are
likely to engage in behaviour if they believe that it will lead to good outcomes, if
other people significant to them will approve, and if they have the necessary
resources and opportunity to perform the behaviour.
TRA and TPB are widely used in family planning research, particularly in relation
to condom use for HIV/AIDS prevention. A number of studies show support for
TRA application in relation to the piU, condoms, lUDs and other confraceptive
metiiods (Jaccard and Davidson 1972; McCarty 1981; Doll and Orth 1993).
Perceived behavioural confrol was found to be a significant predictor of intention to
43
Health workers
often are cited as the positive referents, because they encourage use of family
planning. On the other hand, in studies conducted in Indonesia and Vietnam, they
were shown to have a negative influence on the choice of method, health providers
often advised people not to use the pill due to their limited knowledge or to
misunderstanding (Knodel et al.l995; Huh 1999).
Although TRA/TPB have shown thefr success in family planning appUcation
research, some critiques are worth discussion Sheppard et al. (1988) reports that the
model is often appUed within the rigid framework of the ideal model, where it
44
impUes that individuals have confrol over the behaviour. However, this excludes
socio-economic and poUtical factors that are potentially influential on behaviour.
Conner and Norman (1995) also believe that the model deals with perceptions of
confrol and not with actual confrol issues themselves.
2 .5. Self- efflcacy and contraceptive behaviour
Self-efficacy appears in many theories of behaviour change and it presents the key
constmct of Social cognitive theory. Glanz et al. (1997:164) defines self-efficacy
to be "the confidence a person feels about performing a behaviour". According to
Bandura (1986), self-efficacy is the most important prerequisite for behavioural
change.
capability are also more persistent in the face of difficulties than those with lower
efficacy. The measures should include three factors: the behaviour, the specific
situation and the timeframe.These should be measured against levels of the various
task demands.
Perceived self-efficacy has been studied with respect to prevention of unwanted
pregnancies and confraceptive use. Levinson (1986) examined the relationship
between teenage females' confraceptive behaviour and self-efficacy; those with
high self-efficacy scores were sexually responsible and used confraception more
consistently than did thefr low self-efficacy counterparts.
In another study,
48
The self-efficacy constmct was adapted from social cognitive theory (Bandura
1986) and reflects a person's confidence in completing the health behaviour change.
The theory postulates that confidence in one's abiUty to perform a specific
behaviour is sfrongly related to one's actual abiUty to perform, and it has been
associated with stage of change. The situational self-efficacy measures reflect the
confidence of the individual not to engage in a specific behaviour across a series of
difficuh situations (VeUcer et al. 2000)
In TTM, fransitions between adjacent stages are the dependent variables, and the
other constmcts are assumed to influence these transitions, the independent
variables (Norman et al. 2000).
A vast majority of TTM studies have used a cross-sectional design, which allows
for comparison of people in different stages. The pros and cons and self-efficacy
often have proven to be important predictors in certain stages (Weinstein et al.
1998). In addition to that, this design also allows for prediction and for particular
discontinuity pattems (Norman et al. 2000).
According to Norman et al. (2000), few TTM studies have used a longitudinal
design. Longitudinal design can be used to examine the sequences of fransition
through the stages and test whether different theoretically relevant variables predict
stage fransition among people in different baseline stages. Experimental design
allows for testing whether the stage-matched intervention is more effective than an
intervention that is not stage-matched. Studies using TTM for designing stagematched intervention have shown significant success in moving people to the next
stage in the sequence (Weinstein et al. 1998). However, Norman et al. (2000) raise
concems about the evidences of success in some stage-matched intervention studies.
Thus, TTM has been extensively appUed to explain a variety of health behaviours
including confraceptive behaviour. The theory has also been subject to criticism
such as the lack of standardization of staging algorithms and lack of evidence of
success of stage-matched intervention. Nevertheless, TTM, and particularly SOC,
has shown an important paradigm shift in health behaviour by indicating that
behaviour change should not be regarded as a dichotomy but as a process (Norman
49
et al. 2000). ft is a valuable resource for providers of family planning programs who
need to design behavioural change intervention.
2.6.2. Reasons for investigating men's readiness to accept lUD
Confraceptive behaviour involves the decision to adopt and continue using a
confraceptive method (Heise 1997). The domain of confraceptive behaviour is broad
and complex. The Uterature review provides evidence of personal, interpersonal,
environmental, socio-cognitive and programmatic influences on confraceptive
acceptance. Confraceptive behaviour is also highly variable across situations and
contexts, especially as confraceptive need changes during the Ufe cycle. Some
confraceptive methods, such as condoms or the pill, may be suitable for young
unmarried people. Other reversible methods, such as the lUD, may be more
appropriate for married couples with at least one child, and who have low risk of
getting STD/HTV/AIDS.
Choice of method is influenced by the specific circumstances of each couple, by
their social and cultural environment, and by the national and local family planning
program and policies (Shah 1994). Some degree of choice, even within a rather
narrow range of choice seems to be desirable. In relation to rural Vietnam, most
confraceptive use takes places within marriage, exposure to STD/HTV/AIDS is low
and few exframarital relationships occur. Then, the lUD use is a good choice for
many people. It is highly effective, convenient and free of charge. Some side effects
are reported, but they are marginal in relation to having an unwanted pregnancy and
abortion.
Many studies show that the husband's approval appears to be a major determinant
of confraceptive uptake in developing countries and more effective male targeting
may be necessary for maintaining the success of family planning in the future
(Kamal 2000). In a majority of cases in Vietnam, wives take the responsibility to
practice confraception, but the husbands are the main decision makers (Johansson et
al. 1998b; Mai and Montague 1998).
50
The precontemplation stage includes those (couples) who do not use the lUD
and do not intend to start in the next six months.
The contemplation stage includes those (couples) who do not use the lUD but
intend to start in the next six months.
The preparation stage includes those (couples) who have a sfrong intention to
change in the near future (often within the next 30 days) and may participate in
discussion or seek information related to lUD use.
The action stage includes those (couples) who have accepted the lUD for
confraception.
The maintenance stage includes those (couples) who have accepted and used
lUD for confraception for six months or longer.
51
The pros and cons and self-efficacy for confraception in general have been reported
and validated from other sttidies (Galavotti et al. 1995; Grimley et al. 1995; Schnell
et al. 1996; Stark et al. 1998). These constmcts have been developed and vaUdated
for lUD use in rural Vietnam in this study (Chapter 4).
The SOC identifies the stage, where a person is in the change process. According to
the TTM, it is best to provide an intervention that targets the stage to meet the
persons' current need. Prochaska et al. (1994) recommended that the intervention
have a cognitive focus when targeting people in the early stages, so that it is more
likely to help them see that the advantages outweigh the disadvantages. In confrast,
for people in preparation and higher stages, behavioural sfrategies are the best
approach.
The SOC constmct has been applied as a component of intervention studies in a few
confraceptive behaviour trials, mainly for HIV/AIDS prevention (Cabral et al. 1996;
Collins et al. 1999; Fishbein et al. 1999; Malotte et al. 2000). The evidence from
these studies shows that the optimal sfrategy for moving people from the early stage
is to target both cognitive and action-oriented variables like self-efficacy (Malotte et
al. 2000).
A cross-sectional design allows for estimating differences between people in
different stages, which justifies the application of the model. The use of
experimental design with stage-targeted intervention helps to measure the
intervention's effect in progressing people along the stages of change. The purpose
of the current study was therefore to examine the rural Vietnamese men's readiness
to accept lUD for confraception, and also to determine the variables (predictors) that
discriminate significantly between men in different stages, to design a stagetargeted intervention based on predictors, and to evaluate the intervention in
promoting men's readiness to accept the lUD.
Seven classes of variables identified by Prochaska and Diclemente (1983) are
defined in the following manner:
1. Pros for confraception in general are the perceived benefits of general family
planning (confraception)
52
2. Cons for confraception in general are the perceived costs of general family
planning (confraception)
3. Pros for lUD use are the perceived benefits of lUD use.
4. Cons for lUD use are the perceived costs of lUD use.
5. Self-efficacy for contraception in general is the confidence of a person to abstain
from sexual intercourse or to use a confraceptive method in specific situations.
6. Self-efficacy for lUD use is a man's confidence about convincing (persuading)
his wife to adopt lUD use or to continue lUD use.
7. SOC for lUD use, identified by the staging algorithm, the stages of readiness to
accept the lUD for confraception. These terms (stages of change for lUD and
stages of readiness to accept lUD for confraception) wiU be used
interchangeably throughout the thesis.
Two other variables adapted from Green (1994) for additional explanation of men's
readiness to accept lUD are defined as:
1. Men's confraceptive knowledge is referred to the spontaneous recall of modem
andfraditionalconfraceptive methods.
2. Men's communication on family planning with wives and others is referred to
the level (high or low) of communication.
With these specific variables, the underlying hypothesis of the study is that greater
knowledge of a confraceptive method, more favourable attitude towards of family
planning and the lUD method, and more effective communication on family
planning will lead to a higher stage of readiness to accept lUD use, then the
intervention targeted to stage of change will facilitate the positive progress towards
stages of change for lUD use.
2.7. Research goal, aims and questions
Research goal
The research goal for this study is to identify methods by which targeted health
behavioural change programs might be best developed and implemented for men
that will result in increased acceptance of modem confraception.
53
Research aims
To achieve this goal, tiie aims of this study of men in rural communes Vietnam are
to:
1. identify potential modifiable factors of men that relate to acceptance of
confraception; and
2. design and evaluate a targeted health behavioural change intervention to
promote men's acceptance of confraception.
Research questions
1. To identify measures of rural Vietnamese men's readiness to accept lUD for
confraception.
2. To identify the potential modifiable factors in each stage of men's readiness to
accept lUD for confraception.
3. To test whether an intervention targeted at stage of readiness will result in an
increase in men's readiness to accept lUD for contraception.
54
55
56
Population: Vietnam is the twelfth most populous country in the world with a
population of 76.3 milUon and the second most populous country in South East
Asia, after Indonesia. The population is concenfrated mainly in the two large river
deltas. The Mekong River Delta is the most populous region with a population of
16.1 million (21 percent of the total population) and the Red River Delta has a
population of 14.8 million (19 percent of the total). The population density overall
is not particularly high at 231 people/km^, but it is very high in the river deltas. The
people are mainly ethnic Vietnamese (Kinh), with ethnic minorities in the westem
part and cenfral highlands. The country has undergone dramatic demographic
changes over the past ten years. The population growth rate has fallen rapidly, and
fertility and mortality rates have decreased to levels comparable to more developed
South East Asian countries (UNDP 1999).
Fertility: The fertility levels in Vietnam have steadily declined in the past two
decades (Phai et al. 1996; NCPFP 1999). The recent DHS showed that the fertility
levels declined in Vietnam, from 4.0 births per woman in 1987 to 2.3 births in 1997
(NCPFP 1999). The decline in the fertility rate is due to both a drop in the fertility
level and fertiUty pattem changes (NCPFP 1999). The peak fertility now occurs in
the 20-24 years female age group compared to its previous occurrence in the group
aged 25-29 years. The age-specific fertility rate has also changed significantly.
Fertility declines are proportionately greater for women aged 25 years and older.
This arises during the process of fertility fransition, when older women, who have
reached their desfred family size, often make more effort to limit births than do
young women who are in the process of achieving the desfred number of children
(Phai et al. 1996; NCPFP 1999).
Education: Vietnam's educational indicators are much better than would be
expected for a country at its level of income per capita. One contributing factor to
the generally impressive health indicators is the high literacy rate. Nine out of ten
adults are now literate. Primary and secondary school enrolment rates have
increased, offering more hope and opportunity for the nation's children (UNDP
1999).
57
Gender relations in the family: Vietnamese family ideology and norms for social
relations reflect the Confucian fradition. The fraditional Vietnamese family is
patriarchal, patrilmeal, and patrilocal, often with two or four generations under one
roof A strictly hierarchical order reigns, the younger generation obeying the older,
women obeying men (Johansson 1998).
Men have higher status than women, and sons are valued more highly than
daughters. A traditional Vietnamese woman is governed by three basic tenets from
Confucianism.
First, she must submit to her father, next she must obey her
husband, and then if widowed, obey her eldest son. Ideally, she is soft spoken, and
above reproach for her moral conduct (UNDP 1995; Johansson 1998).
Tasks are divided along gender lines in Vietnam; the father typically works outside
the home and the mother/wife takes care of the children and manages the household.
The father also leads the family in ancestor worship (UNDP 1995).
Son preference is sfrong in Vietnam (Haughton and Haughton 1995) because only
men can perform the rites of ancestor worship. However, there is no marked
difference in the freatment of young boys and girls once they are bom in terms of
access to health and education (UNDP 1995).
In the Vietnamese cultures, it is clear that women have the subordinate role.
However, after independence, in 1946, the first constitution states that 'women are
equal to men and enjoy all civil rights' (UNDP 1995). The government of Vietnam
has sfrong policies on gender equality. A number of reforms and laws which forbid
outdated customs like polygamy and forced marriages and formally guarantee equal
rights between women and men in all fields have been stipulated. Different laws,
decrees and resolutions including marriage and family laws to protect the status of
women have been issued. The most recent decree. Number 37, states that all levels
of government and party should have at least 20 percent of elected positions filled
by women. The Vietnam Women's Union and the Committee for the Advancement
of Women are the main organizations working for women's improvement (UNDP
1995).
58
After Doi Moi, Vietnamese women have shared the improved living standards of
the majority. Aduft literacy rate is 88.7 for women and 95.8 percent for men.
However, the difference between men and women over 25 years of age in literacy is
15 percent. Women constitute 52 percent of the labour force, with the highest
proportion in agriculture (71 percent). The average percentage of female-headed
households in Vietnam is 32 percent and nearly 23 percent of the population live in
female-headed households (Kirjavainen 2000). Confrary to the situation in many
countries, female-headed households in Vietnam are not worse off than maleheaded households in terms of living conditions and per capita daily expenditure
(Loi 1996).
However, there is a large variation within and between regions with regard to
women's status. The education and health of mral and highlands women are at a
much lower standard than that of urban women (Johansson 1998). The birth rates
and mortality in some mral areas and highlands are very high and form a major
burden in the lives of women due to their economic, socio-cultural and even
linguistic isolation, and lack of services and infrastmctures (NCPFP 1999).
Although men now tend to be more involved in household work, women still have
much more responsibility for work related to reproduction. Due to the influence of
feudalistic thinking, women have historically suffered disadvantage: lack of
occupational opportunities, poor reproductive health, as well as the overall malecenfred attitude of society (UNDP 1995).
Women's unequal position in decision-making processes and access to resources has
deleterious effect on their reproductive health. According to Ministry of Health
statistics (1999), approximately, one-third of all births are unwanted at the time of
pregnancy. Abortion is common and alarmingly rising and accounts for about 40
per cent of all pregnancies. Some reports mention that about one-thfrd of all
abortions are among yoimg, unmarried women (MOH 1997). In many districts of
the country, there are more abortions than live births.
Gender inequality in freatment is also a problem in Vietnam (UN 1999). Women
tend to receive poor care of and support for their reproductive health. The barriers
to better care include geography, language and culture, attitudes and gender of
59
for
technical
questions of
People's Health Care and Protection for 1996-2000', does not provide a clear
description of purpose nor clear measures to address inequality, quality, fimding and
other issues identified as problems. Consequently, efforts to address these issues
have not been particularly effective (UN 1999).
The literature on utilization of health services indicates that there is a general under
utilization of community health services. The utilization of curative services in
1998 was only 40 percent (MOH 1999). The quaUty of health services delivery
60
incentives and penalties for those violating the policy. Different kinds of penalties
were implemented. Government cadres who had a third child could lose thefr jobs.
For the rural peasants, the fines for having a third baby ranged from about one to
61
three months earnings. But there was large variation in implementing the policies
among provinces and districts (Goodkind 1995).
hi January 1993, the government for the ffrst time approved a resolution on
population and family planning. This resolution was a response to the concern of
excessive population growth and its consequent social, economic and ecological
problems. The resolution endorsed that each family should have nor more than one
or two children in order to lower fertility and achieve population stabiUzation. To
implement the resolution, the Prime Minister approved an official and
comprehensive population and family planning sfrategy to the year 2000 in June
1993 (Phai et al. 1996).
Structure of family planning programs
Since its estabUshment, NCPFP and MOH have closely collaborated in the deUvery
of the family planning program in Vietnam. NCPFP is a ministerial organization
responsible for initiating family planning policies, developing national sfrategies
and coordinating all program activities.
ministries as well as from mass organizations such as Women's Union and Youth's
Union. In 1995, the NCPFP established a network of committees at lower levels,
i.e., the provincial, district and commune levels (Knodel et al. 1995).
The MOH is responsible for the delivery of family planning services through its
network of public health services. Within MOH, responsibility for family planning
services is assigned to the department of Matemal and Child Health and Family
Planning (MCH/FP). Other ministries and mass organizations (most notably the
Women's Union) are involved as the role of educating, promoting and encouraging
the use of family planning (Knodel et al. 1995).
Contraceptive methods use in Vietnam
The CPR in Vietnam is high at 73 percent and has been rising for the last ten years
(NCPFP 1999). However, promotion of confraceptive use is oriented primarily
towards married women. The lUD still is the main method accounting for 38.4
62
percent of current use. Other methods like condoms and the pill account for only
about 5 to 6 percent. A very small proportion of people use male sterilization or
injectables (less than 1.0 percent)
Different sources of confraceptives are available, but the public sector remains the
major source in Vietnam. Confraceptives are available through the commune health
clinics, hospitals, private clinics and fanily planning collaborators. Government is
dominant in providing lUDs and steriUzation. Besides that, confraceptives also are
available at pharmacies, and through some NGOs (UNFPA 2000).
In 1993, UNFPA was the primary, ahnost exclusive, source of public sector
confraceptives. The MOH was responsible for receipt, storage and distribution of
confraceptives. Since 1996, this task has shifted from MOH to NCPFP, supported
by the Population and Family Health project. Another notable change since 1993 is
the expansion of community based distribution of condoms and oral pills. This has
significantly contributed to the increasing prevalence of all confraceptive methods
(UNFPA 2000).
Review of prior studies on family planning program and contraceptive use
In the last two decades, information about reproductive health knowledge, attitudes
and behaviour has been collected systematically by DHS. The first Vietnam DHS
was conducted in 1988 independently of the intemational standard DHS although it
used a modified and considerably abbreviated version of the typical DHS
questionnaire (Phai et al. 1996).
questionnaires DHS have used in other developing countries. The sample consisted
of ahnost 10,500 ever-married women aged 15-49years taken across all 53
provinces of the country. More recently the third Vietnam DHS was conducted in
1997 from June to October under subconfract to Future Group Intemational and
63
General Statistical Office. The sample consisted of 5664 ever-married women aged
15-49 years from across all provinces of the country (NCPFP 1999).
Other surveys related to population issues have also been undertaken during the last
few years, but they have been less comprehensive in coverage or in their freatment
of reproductive health behaviour and attitudes than the three surveys mentioned
above. Nevertheless, valuable information about fertiUty and other family planning
issues has been derived from the reproductive health survey in 1995 (VNRHS 95),
which was conducted in five provinces by NCPFP and Deutsche Gesselschaft fiir
Technische Zusammennarbeit (NCPFP and GTZ 1995). The sample consisted of
6871 women aged 15-49 years. In addition, other smaU cross-sectional health
surveys have been conducted by Mai and Montague (1998); Care Intemational and
MOH (1997); and Mai et al. (2001).
3.2.4. Study population
Hai Phong is in the northeast of Vietnam and has a population of over 1.7 milUon.
The province is divided into 13 districts (nine rural and four urban) and 217
communes. An Hai is a mral district, with a total population of 219,150 people. The
district has one district hospital, four inter-communal clinics and 23 commune
health cenfres. Family planning services such as lUD provision and sterilisation are
offered free of charge at the district hospital, the inter-communal clinics and at
commune health cenfres. Other methods such as condoms or the pill are available
for purchase in dmgstores or clinics. Twelve out of 23 commune health cenfres
reported no family with a third child in thefr coverage areas. Each village has a
village health worker. The CPR is 75 percent with dominance of lUDs (60.0
percent), followed by condom (4.5 percent), the pill (2.5 percent) and a small usage
of other methods like vasectomy or female tubectomy (An Hai District Health
Cenfre 2001).
Two communes (Quoc Tuan and An Hong) were randomly selected from eight rural
communes in An Hai District, where the majority of people earn thefr living by
farming. A small proportion of the population live by fishing and the manufacture
of seafood products. The rest live by other jobs (flower growing, handicrafts, etc).
64
The lUD rate is 60.0 percent at Quoc Tuan and 54.0 percent at An Hong commune
(An Hai District Health Centre 2001). The map of two communes is in Appendix A.
3.3. Measures
Four main measures were included in the study: (1) confraceptive knowledge, (2)
communication on family planning issues, (3) social cognitive factors and (4) stage
of chMige for lUD use. The justification of their inclusion in the study was provided
at previous section (section 2.6.3).
Measures of confraceptive knowledge were adopted from standard DHS to measure
spontaneous and prompted recall of modem and fraditional methods (Ezeh et al.
1996).
Measures of communication on family planning issues were adopted from Mai and
Montague et al (1998) to investigate men's communication on family planning
issues with wives and other people.
Decisional balance and self-efficacy for confraception in general were adopted from
Grimley et al. (1995) and Galavotti et al. (1995). The scales of decisional balance
and self-efficacy for lUD use were developed and validated in the pilot study (Ha et
al. in press. Appendix L).
Stage of men's readiness to accept lUDs was adopted from Grimley and Lee (1997)
and validated in the pilot study (Ha et al. in press).
3.4. Implementation
3.4.1. Funding
Funding for the research was provided by a grant from the China Medical Board of
New York to the Hanoi School of PubUc Healtii
65
3.4.5. Consent
Fully informed consent was obtained from all men participating in the research and
as the sensitive nature of the topic raised the possibiUty that the interview process
may overwhelm some participants, men were assured that they could withdraw
66
consent at any time (see Appendix C). All survey information was confidential, and
the men were assured that the research would not affect thefr ongoing relationship
with local health workers and local authorities. Each participant was assigned an
identification number (ID) and all data were kept in a lockedfiUmgcabinet and on a
password-protected computer.
3.4.6. Data collection
Data was collected in face-to-face interviews. The literature suggests that the
reliability and vaUdity of self-reports is contentious and it is acknowledged that
fictitious reports can and do occur (May 1993). Underestimation of the tme extent
of the problem was considered to be the most likely prevalent in this study, rather
than overestimation.
3.4.7. Training
Prior to beginning the data collection of each study phase, the interviewers
undertook fraining at the An Hai district health cenfre. Face-to-face interview
requires careful fraining. Training should give interviewers experience of and
confidence to interview people. The fraining should ensure that interviewers can
read fluently, speak clearly and are able to answer the interviewees' questions (De
Vaus 1995). The number of interviewers depends on the nature of the study. The
pilot study involved only four interviewers while the baseline and posttest involved
12 interviewers. A manual for data collection was developed by the researcher,
which explained how to ask and record the answer for each question in the
questionnaire (Appendix D).
Interviewers, both male and female, were local health workers. Those people were
chosen because they were familiar by the respondents. During two days of fraining,
the interviewers practiced adminisfration of the questionnaire by completing it
themselves and through role-playing.
67
scales were in the same direction. The next step was to detennine through
correlation matrix the association among variables. If the correlation was higher
than 0.9, then the items were excludedfromthe analysis
The reUability was measured by internal consistency reliability in all three studies
(pilot, baseline and posttest), however only test-retest reUability was reported for the
pilot study.
3.6. Data analysis
Three sets of analyses were undertaken using the SPSS: (1) Descriptive analysis; (2)
Bivariate analysis and (3) Multivariate analysis.
The detailed data analyses of the pilot, baseline and posttest study are provided for
each study, respectively.
69
70
FGD participants were asked about the practice of confraception with reference to
the lUD method, their perceived benefits and costs of family planning and thefr
perspective on male involvement in family planning. They also were asked to
describe the factors related to lUD use such as benefits (one time insertion, free of
charge) and consfraints, as well as the situations in which women found it hard to
use lUDs (abdominal pain, bleeding). The interviews and FGDs were franscribed
and content was analysed for development of the quantitative survey instrument.
The content analysis identified specific issues for TTM constmcts, and the
information was used to develop the Likert-type scale questionnaire.
Scales were developed to elicit information concerning pros and cons (benefits and
costs) and self-efficacy for lUD use (confidence in dealing with lUDs).
4.3. Pilot survey
A cross-sectional survey by face-to-face interview was carried out after the
qualitative phase.
A total of 201 men who satisfied screening criteria were selected for the study. The
screening criteria were described in Chapter 3 (Section 3.4.4).
The participants' mean age was 35.7 years, with a Uttle over one-thfrd of men in the
sample less than 35 years old. Over two-thirds (69.7 percent) of the men were
working in the local agricultural fields.
4.4. Test- retest
A total of 29 questionnaires were re-tested in an actual field sittiation in two weeks.
Test-retest is a method to check on the reUability of questions, to ask the same
people the same questions after an interval of two weeks and calculate the
correlation. If the test-retest coefficient is greater than 0.5; then the question is
assumed reliable (Sfreiner and Norman 1998). However, the test-retest method is
not always a good test because it relates to length of time as well as the participant's
71
memory (De Vaus 1995). Therefore, it can artificially mflate the apparent reUabiUty
of the question.
4.5. Measures
The pilot questionnafre consisted of two parts: basic socio-demographic
characteristics; and items assessing the three main TTM constmcts, i.e. stages of
change for lUD use, pros and cons and self-efficacy. The staging algorithm for lUD
use was adapted from Grimley and Lee (1997); and the decisional balance and selfefficacy for confraception in general were adapted from Grimley et al. (1995).
These items were developed and validated in other studies (Galavotti et all995 and
Stark etal. 1998).
Items for measuring self-efficacy and decisional balance for lUD use were
developed using the results of the qualitative phase. Items for the scales that
originated from scales developed in English were franslated to Vietnamese.
Following the content validation by experts, each item was back franslated to
English to ensure it contained the same meaning. All items were reviewed by family
planning experts and health behavioural experts for its content. Additionally, these
items also were reviewed by family planning health workers and married men in the
two rural communes. An Hong and Quoc Tuan, Vietnam.
4.5.1. Stages of change algorithm
To assess readiness to adopt and maintain lUD use as a confraceptive method, a
four-item staging algorithm was used. These items have been found to be reliable in
previous studies (Grimley and Lee 1997). The first question asked participants
whether they used an lUD or not: 'Do you currently use lUD method for pregnancy
prevention?' (Yes/No). People who currently used lUDs were asked: 'How long
have you been using an lUD?', and the dichotomous answers were: 'less than 6
months' or 'more than 6 months'. Those who were cunently not using an lUD
were asked the question: 'Are you (your spouse) thinking about using an lUD in the
next 6 months?' (Yes/No). Those who answered 'yes' were asked: 'Do you plan to
use a lUD in the next 30 days?' (Yes/No). Responses to these questions were used
72
to classify the individual into one of five stages of change. The staging algorithm is
shown in Table 4.1.
Table 4.1. Staging algorithm for lUD use
Algorithm
Stage
Ql. Is your wife currently using an Yes->^Q2
No -^ Q 3
lUD?
Q2. How long has your wife been using
an lUD?
73
4.7. Results
4.7.1. Factor analysis of decisional balance scale
The original decisional balance scale was divided into four sub-scales: pros/cons for
confraception in general, and pros/cons for lUD use. These items are shown in
Table 4.2.
75
Table 4.2.Items to assess pros and cons for contraception in general and lUD
use
Item
No
Contents of constructs
Pros for contraception in general
1.
2.
3.
4.
5.
6.
7.
Some confraceptive method like condoms and the oral pill requfre
preparation for sex
8.
9.
10.
11.
12.
13.
14.
15.
16.
76
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
To explore the underlying stmcture of items assessing decisional balance (pros and
cons for confraception in general and pros and cons for lUD use), all 27 items were
factor analysed using principal component analysis with varimax rotation. A
correlation matrix was computed for all variables. This showed a large number of
Pearson's coefficients with values greater than 0.3, indicating evidence of
reasonable correlations between variables (Tabachnix and Fidell 1996).
The Bartlett test of sphericity produced a significance level of <0.001 indicating that
the data were from a multivariate normal distribution (Tabachnix and Fidell 1996).
The Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy was 0.81,
indicating an acceptable index for comparing the magnitude of the observed
correlation coefficients to the magnitude of the partial correlation coefficients
(Kaiser 1974).
Using the default settings for principal component analysis, six factors having
eigenvalues greater than one and accounting for 57.6 percent of total variance were
exfracted from the correlation matrix. These factors were rotated to orthogonal
77
simple stmcture. The pattem of rotated factor loading did not fully support the
items groupings hypothesized earUer. Two new factors were found, and several
items tended to correlate higher with different factors than had been hypothesized.
Results of the rotated factor solution are shown in Appendix E.
Factor 1: First factor accounted for 20 percent of total variance of scale. Sbi items
emphasizing disadvantages of the lUD remained tiie empfrical core of the cons of
lUD (item 21-26). However, item 26 had a secondary loading on factor 5.
Factor 2: Second factor accounted for 13 percent of total variance of scale. Six
items emphasizing advantages of confraception remained the core of pros of
confraception (item 1-6), with additional item (item 13) originally hypothesized to
be pros of lUD. However, item 13 loaded better on the thfrd factor.
Factor 3: Third factor accounted for 7 percent of total variance of scale. Seven
items emphasizing advantages of lUD remained the core of pros of lUD (item 1319), with additional item (item 20) originally hypothesized to be cons of lUD.
However, item 19 and item 20 had secondary loading on factor 6.
Factor 4: Fourth factor accounted for 6 percent of total variance of scale. Five items
emphasizing the disadvantages of confraception remained the core of cons of lUD
(item 7-10, and item 12) with additional items (item 27) originally hypothesized to
be cons oflUD.
Factor 5: Fifth factor accounted for 5 percent of total variance of scale. Five items
formed a new factor (item 7, 11, 26, 27, 25). However, among these, item 25 and
item 7 had secondary loading on factor 1 and 4, and other two items 7 and 26 also
loaded in another factors (factor 4).
Factor 6: Sixth factor accounted for 4 percent of total variance of scale. Four items
formed a new factor (item 18, 12, 19, 20). However, among these, two items (18,
and 12) had secondary loading on factor 3 and 4, and other two items (19 and 20)
also loaded in other factors (factor 3).
78
Overall, the factor aialj^ic finding suggested that hypothesized item groupings
based on the decisional balance required some modifications. The grouping items
should be quite distinct from each other; e.g., there were no items overlapping
secondary loading (>0.3) on the other factors, and the item that were highly
correlated should be excluded. Using these mles, item 22 was excluded because it
was highly correlated with item 23 (R> 0.95). Items that had secondary loading
were also excluded by keeping 3 or 4 items for each sub-scale, it was found that 14
best fitted items remained in the scale. These items were factor analysed with
varimax rotation. Four factors (eigenvalues >1), which accounted for 65.8 percent
of the total variance were exfracted. Scores on the new items were computed by
averaging unweighted ratings for the individual items within the four factors (Table
4.3).
79
Table 4.3. Retained decisional balance scale items and factor loading
Factor - items
Factor loading
1
0.91
0.90
may
cause
PID
(pelvic
0.80
inflammatory disease)
Factor 2: Pros for contraception in
general
(a-=0.89)
FP helps you to avoid the results of
0.78
unwanted pregnancy
Your wife would not have to worry
about
becoming
pregnant
0.77
if using
confraception
FP helps to limit size of the family
FP
helps
you
to
become
0.73
more
0.69
0.85
pregnancy
lUD may work for a long time
lUD does not require preparation before
0.82
0.67
sexual intercourse
Factor 4: Cons for contraception in
general
(a ^0.61)
FP makes sexual intercourse difficult in
0.81
a family setting
Some FP methods are costiy
0.76
0.66
80
Contents of constructs
No
Self-efficacy for contraception in general
1.
2.
3.
4.
5.
When the wife suffers side effects like nausea, pain etc
Self- efficacy in convincing wife to have lUD inserted
6.
I am confident that I could discuss lUD use with my wife even if she did
not want to.
7.
8.
9.
10.
11.
12.
81
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
82
Items
Loadings
contraception
in general
When your wife gets upset about that
When the wife has to go through too much
0.78
0.52
frouble
0.79
(Cronbach 's
^ ^^. When the wife suffers side effects Uke nausea, pain, 0.76
alpha = 0.75) ^^^
Factor analysis with varimax rotation was carried out separately for items
measuring self-efficacy in convincing wives to use an lUD and in convincing wives
to continue to use an lUD (item 6-14 and item 15-22). The Bartlett test of sphericity
showed a significance level of <0.001 indicating that the data were from a
multivariate normal distribution (Tabachnix and Fidell 1996). The Kaiser-MeyerOlkin (KMO) measure of sampling adequacy was 0.79 and 0.75, respectively,
mdicating an acceptable index for comparing the magnitude of the observed
conelation coefficients to the magnittide of the partial correlation coefficients
(Kaiser 1974).
One factor for each situation was revealed with eigenvalues greater than one and
accounting for 65.0 and 72.0 percent of total variance, respectively. The Cronbach's
alpha coefficients of both sub-scales were between 0.8 and 0.89, illusfrating good
intemal consistency of the scales (Table 4.6).
83
Items
Loadings
84
Test-retest coefficient
0.64
0.38
0.53
0.38
0.4
stage;
about
one-tenth
(10.4
percent)
were
in
the
significant differences of cons for lUD use (F (2, 194) = 3.249; P = 0.04). Post-hoc
Tukey test results indicated that there were significantly higher scores for
respondents in the precontemplation stage compared to tiie action/mamtenance
stage.
confraception in general and for pros for both confraception in general and lUD use.
Figure 1. Standardized scores of pros and cons for contraception in general
and lUD use
53.0 n
5Z0
51.0-
50.0
II)
8 49.0-
48.0
47.0
46.0
45.0
Precontemplation
Contemplation/Preparation
Action/Maintenance
52.0 n
51.0
50.0-
"9.0 ^
*~ 48.0
47.0
46.0-
45.0
Precontemplation
Contemplation/Preparation
Action/Maintenance
87
Pros and cons and self-efficacy for confraception in general and lUD use were
measured by usmg T-scores. Prochaska et al (1994) suggest that prior to the action
stage, the pros and cons should crossover with pros being higher than cons. The
findmgs show that the expected pattem of pros and cons was seen where the pros
and cons intersected at the contemplation/preparation stage. However, in the case
of readiness to accept lUD for confraception, it was the cons and not pros that
varied by stage.
89
The implementation of the pilot survey showed good responses of men in answering
questions, and local health workers were able to approach men at thefr house for
interviews. However, some problems were identified such as: (a) difficulties in
meeting the men during daytime; and (b) poor response if participants were too tfred
or too busy at the time of interview. Sfrategies to overcome these problems
included: (a) to visit in the evening, and if the participants were too tfred or busy,
(b) to make a specific time at which to return.
90
91
92
93
Cronbach's coefficient
0.85
0.60
0.78
0.90
0.78
0.90
5.3.3. Validity
Constmct validity of the measures used for this investigation into men's readiness
to accept lUD for confraception in rural Vietnam was established in the pilot study
(Chapter 4).
95
two- tiiirds of men had wanted the last birth. For logistic regression, it was grouped
as 'last birth wanted' and 'last birth not wanted' regardless of tinting.
The frequency of participants' communication with wives and with others on family
planning issues in the year preceding the survey was categorised as 'no
communication', 'communication from 1-3 times', and 'communication more than
3 times'.
Using Jolly (1976), a composite index was created to study the quantum of men's
communication. The index was based on the overall responses pertaining to the
frequency of communication on the specific items. The 'no communication' was
coded '0'; 'communication from 1-3 times' was coded ' 1 ' ; and 'communication
more than 3 times' was coded '2'. Scores for frequency of commimication with
wives as well as with other people ranged from 0 to 10. Men who scored from 0 to 5
98
were categorized as low communicators, and those who scored from 6-10 were
categorized as high communicators.
Perceived pros and self-efficacy for both confraception in general and lUD use were
scaled as (1) high level (scores from 4-5), and (2) low level (scores from 1-3) and
coded as 1/0. The reverse coding was used for cons for contraception in general and
for lUD use.
Reasons for not practising contraception were categorized as (1) planning to have
next child; (2) difficult to conceive; (3) too old; (4) infertile; (5) too complicated;
(6) side effects; (7) too expensive; (8) religiously unacceptable; (9) infrequent
sexual intercourse; (10) unavailable method; and (11) other. A dummy coding was
used for each category.
5. 5. Data analysis
The data from the questionnafres were entered into EPI-INFO version 6, and then
converted to the Statistical Program for Social Science (SPSS) version 11 for
analysis. Frequencies for all variables were examined for missing, unlikely, or outof-range values and when detected were checked against the original data source.
99
The stage of change for lUD use in two adjacent stages is binary outcomes that
generate the binominal disttibution. The logistic model is appropriate to sttidy tiiese
outcomes as it enables assessment of the probability of movmg to the next stage of
change for lUD use in the sequence, and it is based on tiie nonlinear fimction of tiie
best line^ combination of predictors, with two outcomes expressed as:
Y = e^'/(l+e^)
Where Y is the estimated probability that the ith case (i =1,..., n) is in one of tiie
categories and fi is the usual linear regression equation. The logfransformationof
linear regression equation will create the log of the odds:
Ln[Y/(Y-l)]=A-HlBjXij
So, the odd of (dependent variable)
101
The ratio of subjects to independent variables was assessed to ensure that the ratio
was sufficiently high, usmg the formula N>50-i-8m (where m = number of
independent variables (Tabachnick and Fidell 1996).
Normality, linearity and independence of residuals were also assessed. For this
assessment, univariate residual scatter plots were run. Residuals were also displayed
against each independent variable. Plots were examined for linearity, normality and
outUers and independence residuals. The assumptions are that residuals are
normally distributed, that residuals have a linear relationship, and in an unbiased
model, residuals randomly vary around their mean (zero) and around the mean of
the independent variable.
Multicollinearity and singularity for the independent variables were assessed. These
conditions occur if the independent variables are very highly correlated, i.e., r=0.9
and above (multicollinearity), or perfectly correlated (singularity). This means that
one independent variable is actually a combination of other independent variables.
Bivariate correlations were mn and the Pearson's correlation coefficients were
assessed. The software SPSS 11 was used to check the colUnearity in the regression
analysis. No colUnearity problems were identified for the independent variables.
5.6. Results
This section is organised in three parts. The ffrst part presents information on
participants'
socio-demographic
characteristics,
confraceptive
knowledge,
102
information
included
mformation
on age, education,
(0.6 percent).
Approximately two-thirds (61.8 percent) were farmers, less than one-fifth (17.1
percent) had government or private job, and a very small proportion (2.5 percent)
had other jobs.
103
Number of
Percent (%)
participants
Age (years)
19-24
14
2.2
25-29
101
15.5
30-34
151
23.2
35-39
225
34.6
40-44
146
22.4
45-49
14
2.1
49
7.5
Lower secondary
394
60.5
Upper secondary
167
25.7
Higher education
41
6.3
Government
106
16.3
Private
111
17.1
Agricultural
402
61.8
House work
15
2.3
Otiier
17
2.5
648
99.5
Muong
0.2
Nung
0.2
Mixed
0.1
645
99.1
Buddhism
0.6
CathoUcism
0.2
Otiier
0.1
651
100.0
Education (level)
Primary
Occupation
Ethnicity
Kinh
Religions
No religion
Total
104
105
Number of
Percent (%)
participants
Number of sons
None
169
26.0
One
334
51.3
Two
134
20.6
Three
13
2.0
Four
0.1
None
223
34.3
One
299
45.9
Two
108
16.6
Three
20
3.1
0.1
0.2
One
201
30.9
Two
334
51.3
Three
106
16.3
1.3
No
169
26.0
Yes
482
74.0
None
483
74.2
One
123
18.9
Two
38
5.8
Three
0.9
Four
0.2
651
100.0
Number of daughters
Four
Total number of children
None
Four
Having a son
Abortion
Total
106
Table 5.4 presents the men's fertility intention. The mean of desired children was
2.1 (SD =0.5). The minimum desfred number of children was 1 and the maximum
was 5. Two children were the most frequent number of children that men desfred
(84.6 percent), foUowed by three children (8.4 percent). Few men wanted to have
only one child (4.9 percent), and fewer wanted to have four or five children (about 1
percent). None wanted to be childless.
Mean of desired sons was 1.1 (SD = 0.38), and the number of desired sons ranged
from 0 to 3. On comparison, the number of desfred daughters varied from 0 to 2
only, with mean was 1.01(SD = 0.3). A majority of men desfred to have one son
(88.6 percent) or one daughter (90.8 percent). Few of them desired to have two sons
(7.7 percent) or two daughters (4.9 percent) and fewer wanted to have three sons
(1.2 percent).
More than two-thfrds of men did not want to have any more children (70.8 percent).
Approximately one-fourth expressed the desire to have children either in the next
two years (6.6 percent) or later than two years (19.0 percent); a small proportion
was uncertain of time (2.3 percent).
With regards to last birth intention, over two-thirds of men (73.1 percent) reported
the last birth was wanted. About one-tenth (11.1 percent) stated last birth occurred
at an undesired time, and 15.8 percent indicated it was unwanted. Thus, a total of
26.9 percent of men expressed their unmet need for confraception for the family's
most recent birth.
107
Number of participants
Percent (%)
32
4.9
Two children
551
84.6
Three children
55
8.4
Four children
1.2
Five children
0.9
16
2.5
One
577
88.6
Two
50
7.7
Three
1.2
None
28
4.3
One
591
90.8
Two
32
4.9
461
70.8
43
6.6
124
19.0
15
2.3
1.3
Wanted then
476
73.1
Missed time
72
11.1
103
15.8
651
100.0
FertiUty intention
Wanted no more
Wanted within two years
Wanted later
Not sure about time
Do not know
Last birth intention
Unwanted
Total
108
Percent
68.1
0.7
63.5
4.6
109
Ugation (9.0 and 8.9 percent, respectively). Only 1.0 percent of men spontaneously
recalled injectable metiiods. Forfraditionalmethods, less tiian quarter of tiie men
had heard about either abstmence (23.0 percent) or withdrawal (23.7 percent).
Table 5. 6. Percentage distribution of contraceptive knowledge
Methods
Spontaneous
Prompted
Total
(N=651)
(%)
(%)
knowledge (%)
Any method
96.8
83.9
99.4
Modern methods
94.0
68.4
99.2
Pills
44.3
37.9
74.4
Condoms
72.9
27.1
93.5
lUD
84.3
17.6
96.3
Vasectomy
9.0
39.7
42.1
Tubal ligation
8.9
40.3
42.4
Injectable
1.0
0.9
1.7
38.4
40.7
78.1
Periodic abstinence
23.0
19.2
38.4
Withdrawal
23.7
37.7
54.4
0.3
0.4
0.6
Traditional methods
Other
110
Percent (%)
Radio
81.1
Television
92.6
Newspapers
55.2
Billboard
24.3
Health workers
52.2
36.6
Friends/relatives
21.9
Women's Union
25.8
Wife
48.8
Other
.3
Percent (%)
(N=299)
Side effects
71.2
Effectiveness
55.9
Indication
58.9
Instruction
36.1
No need
0.7
111
The
questions were designed to investigate the communication pattem (1) with wives
and (2) with other people (e.g., fiiends, family planning collaborators. Women's
Union members, people's commune committee and relatives) in the year preceding
the survey.
Table 5. 9. Couple's communication on family planning issues
Communication pattern
Number of participants
Percent (%)
196
30.1
1-3 times
87
13.4
> 3 times
368
56.5
408
62.7
1-3 times
72
11.1
> 3 times
171
26.2
562
86.3
1-3 times
40
6.1
> 3 times
49
7.6
No
397
61.0
1-3 times
105
16.1
> 3 times
149
22.9
473
72.7
1-3 times
62
9.5
> 3 times
116
17.8
651
100.0
Total
112
Percent (%)
Number
With fiiends
431
66.2
1-3 times
86
13.2
> 3 times
134
20.6
587
90.2
1-3 times
38
5.8
> 3 times
26
4.0
604
92.8
1-3 times
26
4.0
> 3 times
21
3.2
632
97.1
1-3 times
1.2
> 3 times
11
1.7
509
78.2
1-3 times
71
10.9
> 3 times
71
10.9
651
100.0
No
With relatives
No
Total
Table 5.10 shows the frequency of men's communication with others on family
planmng. More tiian two-tiiirds of men did not communicate witii otiier people on
113
family planning issues. The majority of them (90 percent) did not communicate
with family planning collaborators. Women's Union members, and the people's
commune committee. A small proportion of men (less than 5 percent)
communicated with these groups eitiier from 1 - 3 times or more than 3 times.
Participants had more frequent communication on family planning issues with
fiiends than relatives (33.8 vs. 21.8 percent). Approximately one-fiftii (20.6 percent)
communicated more than 3 times and one-tenth (13.2 percent) communicated 1- 3
times in the previous year with fiiends. In comparison, about one-tenth (10.9
percent) communicated 1-3 times and more than 3 times in the previous year with
relatives.
5.6. 4. Social cognitive factors
Table 5.11 presents scores of social cognitive factors on decisional balance and selfefficacy by means and standard deviation (SD). Generally, participants understood
the benefits of confraception and lUD use. The pros for confraception in general
consisted of 4 items, and scores varied from 8 to 20 with a mean of 17.0 (SD=1.7),
and the pros for lUD use consisted of 3 items with score's ranging from 6 to 15 and
meanofl2.2(SD=1.4).
Table 5.11. Percentage distribution of social cognitive scales
Social cognitive scales
Means
Standard
Deviation
17.0
1.7
6.4
1.2
12.2
1.4
11.9
3.8
19.4
2.6
15.0
2.9
114
Not many participants expressed concems for confraception. Cons for confraception
in general scored low. The scale consisted of 3 hems and scores varied from 3 to 13
with mean of 6.4 (SD=1.2). In confrast, the cons for the lUD held more concerns for
many participants. The scale consisted of 4 items and scores varied from 4 to 20
witii mean of 11.9 (SD=3.8).
Men expressed high self-confidence (self-efficacy) for confraception in general as
weU as for the lUD use. The self-efficacy for contraception in general consisted of 5
items, and the scores varied from 5 to 25 with mean of 19.4 (SD=2.6). The selfefficacy for the lUD consisted of 4 items and scores varied from 8 to 20 with mean
ofl5.0(SD=2.9).
5.6. 5. Practice of contraceptive methods
Table 5.12 presents the level of current use and ever-use of confraceptive methods.
A majority of men (96.9 percent) reported current use of at least one confraceptive
method; only 3.1 percent had not used any method. The lUD was the main method,
which accounted for 62.1 percent and the next was withdrawal (27.9 percent). It
can be seen that the use offraditionalmethods was relatively high (43.2 percent).
Table 5.12. Percentage distribution of contraceptive use
Contraceptive methods (N=651)
Use of confraceptive methods
Current use
96.9
Pill
0.2
Condoms
4.6
lUD
62.1
Abstinence
15.3
Withdrawal
27.9
3.1
Those who were not using any confraceptive methods were asked about the reasons
for not practicing and the results are presented in Table 5.13. The major reason was
115
irregularity of sexual mtercourse (26.3 percent) and fear of side effects (21.1
percent). Financial cost and nonavailability were insignificant (about 5 percent)
Table 5.13. Percentage distribution of reasons for not using contraception
Reasons for not using contraception (N=19) Percent (%)
Difficult to get pregnant
15.8
Too old
10.5
Too much
frouble
5.3
21.1
5.3
26.3
Nonavailability of method
5.3
Other
36.8
Using the staging algorithm, the men were classified into 5 different stages of
readiness to accept the lUD for confraception. Their distribution presented in Table
5.12, shows that men tended to cluster into two groups: precontemplation (29.5
percent) and maintenance (56.4 percent). In confrast, a relatively small number were
classified as being in contemplation (6.1 percent), preparation (4.5 percent) or
action (3.5 percent).
Table 5.14. Percentage distribution of men's stage of change for lUD use
Stages of change for lUD Number (N) Percent (%)
Precontemplation
192
29.5
Contemplation
40
6.1
Preparation
29
4.5
Action
23
3.5
Maintenance
367
56.4
Total
651
100.0
Given the low numbers in some groups, and as a sfrategy for sunplifying the
statistical analysis and increasing sample sizes, the five stages of behaviour change
for lUD use were reduced to three, by combining men in the contemplation and
116
g, 50.0
I 40.0
S 30.0 I 20.0
10.0
0.0
Stage of readiness to accept lUD for contraception
Precontemplation Contemplation/ Preparation Action/Maintenance
117
Age groups
20-24
25-29
30-34
35-39
40-44
45-49
Education
Primary
Lower secondary
Upper secondary
Higher
Occupation
Government
Private
Farming
Other
Parity
One
Two
>Three
Abortion
No
Yes
Having a son
No
Yes
Last birth wanted
No
Yes
Desired children
No
Yes
Number of men
4.4
20.3
23.2
36.2
13.0
2.9
2.3
15.9
24.9
34.1
21.0
1.8
0.2
6.8
58.3
28.6
6.3
4.3
50.7
36.2
8.8
8.5
63.3
22.3
5.9
0.1
18.8
16.7
57.3
7.2
20.3
18.8
52.2
8.7
14.4
16.9
65.6
3.1
0.06
33.3
48.4
18.3
39.1
40.6
20.3
28.5
54.6
16.9
0.2
63.0
37.0
62.3
37.7
81.8
18.2
0.00
28.6
71.4
40.6
59.4
22.1
77.9
0.03
24.0
76.0
21.7
78.3
29.2
70.8
0.2
69.3
30.7
192
65.2
34.8
69
72.6
27.4
390
0.3
The proportion of men in the action/maintenance stage whose wives had not
undergone abortion was significantly higher than those in the lower stages (81.8 vs.
118
62.3 and 63.0 percent; P<0.05). In confrast, the proportion of men in the
action/maintenance stage who had a son in the family was significantiy higher than
those in the contemplation/preparation and the precontemplation stage (77.9 vs.
59.4 and 71.4 percent; P<0.05).
Examining the knowledge and communication by stage of change for lUD use,
there were statistically significant differences between stage of change and
spontaneous recall of modem method, of fraditional methods and communication
with wives (P<0.05) (Table 5.16).
Table 5.16. Relationships between knowledge, communication and stages of
change for lUD use
Characteristics
P value
Recall of modem
method
87.5
95.7
96.9
0.00
Recall of
fraditional method
64.6
52.2
23.1
0.00
Communication
with others
Low
97.4
98.6
94.6
2.6
1.4
5.4
72.9
65.2
82.6
27.1
34.8
17.4
192
69
390
High
Communication
with wives
Low
High
Number of men
0.1
0.00
significmtiy higher than tiiose in the other stages (82.6 vs. 65.2 and 72.9 percent,
respectively).
MANOVA was performed to investigate the differences in decisional balance by
stages of change. Assumption was tested to check for normality, linearity, univariate
and multivariate outUers, homogeneity of variance-covariance matrices, and
multicollinearity. No serious violation was detected.
significant differences between stages of change for lUD use on the combined
dependent variables (pros and cons for lUD use) (F(l,650)=l 1.364; P<0.05). The
one-way ANOVA and Tukey post-hoc test were used to examine these factors by
stage of change for lUD use (Table 5.17). The pros for lUD use increased from
precontemplation to action/maintenance stage and the reversefrendwas reported for
cons for lUD use. The crossover occurred in the contemplation/preparation stage.
Men in the action/maintenance stage reported significantly higher pros for lUD use
than those in the precontemplation stage (51.4 vs. 47.3; F(2, 650)= 11.043; P<0.05).
In confrast, men in the action/maintenance stage scored significantly lower cons for
lUD use than those in the two other stages (47.6 vs. 52.3 and 54.0; F(2, 650)=
31.636; P<0.05).
120
constructs
Pre
Contenq)lation/
Action/
contenq)lation
Preparation
Maintenance
(N=192)
(N=69)
(N=390)
Mean (SD)
Mean (SD)
Mean (SD;
Pros for
confraception
49.6(9.0) 51.3(11.2)
F ratio
P
value
50.0(10.3)
0.753
50.4(10.6)
0.873
0.47
Cons for
confraception
49.3(9.0)
Pros for
47.3(11.0)
49.9(9.5)
51.4(9.3)
11.043
54.0(9.2)
52.3(9.9)
47.6(9.7)
31.636
flJD
49.6(9.2)
0.42
0.00
0.00
0.00
52.7(9.4)
50.3(10.3)
48.6(10.0)
10.747
42.8 (13.0)
50.1 (7.9)
53.5 (5.8)
95.487
0.00
between stages of change for lUD use on the combined dependent variables (selfefficacy for confraception in general and for lUD use) (F(l, 650)=48.696; P<0.05).
Self-efficacy for confraception in general decreased from precontemplation to
action/maintenance stage, while the reverse result was reported for self-efficacy for
lUD use. Results of one-way ANOVA and post-hoc Tukey test showed that men in
the action/maintenance stage reported significantly lower self-efficacy
for
121
In each analysis, the dependent variable was coded to represent being in a higher
stage for lUD use, allowing for easy determination of factors that may be usefiil in
designing interventions to facilitate men's movement from a lower to a higher stage.
For each independent variable in the multivariable model, the logistic regression
analysis assessed the odds of being in one stage, such as contemplation/preparation,
compared to being in the precontemplation stage.
After confroUing for age, occupation and education, the model estimation included
only factors that were found significantly associated with stage of change for lUD
use in the bivariate analysis. These factors included history of having abortion,
having a son in the family, spontaneous recall of modem and fraditional
confraceptive methods, communication with wives and others on family planning
issues, pros and cons for lUD use, self-efficacy for confraception in general and
lUD use. No high correlation (R>0.09) was found among these variables.
A modified backward elimination variable reduction sfrategy was used to remove
any of the additional variables if these variables failed to maintain significance in
the final multivariate models. These final models were obtained in the following
way. First, a series of logistic regression with analyses (precontemplation vs.
contemplation/preparation, and contemplation/preparation vs. action/maintenance),
using a selection of all significant variables from the bivariate analysis Usted above,
122
were conducted. Next, variables that failed to show significance in any of tiiese
logistic models were removed, and the process was repeated usmg a more
conservative p value (Stark et al. 1998). The significance across the model used a
liberal p value of 0.2, 0.1, and fmally 0.05.
The results shown in Table 5.18 presents the predictors of the likelihood of being m
the contemplation/preparation stage compared to those in the precontemplation
stage of change for lUD use after controlling for ages, occupation and education.
Table 5.18 is a compilation of significant (P=0.10) predictors identified at each
stage of the model fitting process. Two models were presented: a full and a reduced
model.
The full model containing nine variables provided a significant fit to the data
(P=0.004). The significant independent variables were (a) recall of fraditional
confraceptive method, and (b) self-efficacy for lUD use.
The full model indicated that six variables, namely having a son, history of
abortion, spontaneous recall of fraditional methods, communication with wives,
pros and cons and self-efficacy for lUD use were all positively associated with stage
of change for lUD use. There was an inverse association between stages of change
for lUD use and spontaneous recall of modem methods and self-efficacy for
confraception in general.
In a reduced model, predictors of being in the contemplation/preparation stage were
modeled against variables from the full model statistically associated (P <0.2) with
the dependent variables. The reduced model was almost as good a fit to the data as
the full model but was, by definition, more parsimonious (Chi-square =36.043,
df=14, P=0.001). Both significant variables in the full model retained their
significance in the reduced model e.g., self-efficacy for lUD use and spontaneous
recall offraditionalmethods.
Men in the contemplation/preparation stage were 1.8 times more likely not to recall
fraditional methods than those in the precontemplation stage (0R=1.8; 95%
confidence interval 1.0 to 3.4; P=0.06), other things being equal.
123
Men in the contemplation/preparation stage were 5.2 times more likely to report
high self-efficacy for lUD than those in the precontemplation stage (OR=5.2; 95%
confidence interval 2.3 to 11.7; P=0.00), other things being equal.
124
o
o
b
>
o
M)
9i
fl
I
u
a
o
u
V
T3
a
o
St
t9i
as
I
00
en
la
0H
as
V
,3 '
OO
n
u
en
^
9
vo ^
a
O
o
a
a
o
fS m
-^
vo
O
<N
in
^ o
88
CN
en
0)
O
O
-^
R
O,
en
t--
ON
0^
in
OO
t--
oo
oo
o
o
rn
^^
(>
4
b
^
^
n
=^
rij
(N
OO
cs
iI
ov
c^
'I
rn
en
as
b
9i
U
a
JO
'S
ct
"a
9i
"a
a
o
U
*-<
tt
a
x
!S
a
o
t
a
CN
I
^ ^ ^ d
^
IT)
ON
es
pa
0.2
"W "o,
o ST
o w
a fc.
a
^ H
r~^
-^
^'
>n
b
c-^
.I
en
,I
t~~rn
T-;
.I
as
I1
in
CN ^ CN O
vo
ia S
u
b
4^
U
U
^ et
-^ o
. * en
'SD (A
es
.Pi4
* *
.18.
ead
NJ
ifi
2es
k.
en
a
4>
o
CO
o
o
13
-^
00
X ^
PLH
>n
CN
The results shown m the Table 5.19 presents tiie predictors of tiie hkelihood of
bemg
in
the
action/mamtenancestage
compared
to
those
in
the
126
Men in the action/maintenance stage were 0.7 times less likely to have wives who
had undergone abortion (OR=0.3; 95% confidence interval 0.2 to 0.6; P=0.00). In
other words, men in the action/maintenance stage were almost 3 times more likely
to have wives who had not undergone abortion.
Men in the action/maintenance stage were 3.8 times more likely to not recall
fraditional confraceptive methods than those in the contemplation/preparation stage
(OR=3.8; 95% confidence interval 2.0 to 6.9; P=0.00). Those people who were
currentiy using lUDs were less likely to recallfraditionalmetiiods tiian those in tiie
lower stages.
Those in the action/maintenance stage were 1.8 times more likely to have low
communication with wives on family planning matters than those n a lower stage
(0R=1.8; 95% confidence interval 0.9 to 3.5; P=0.09). Thus, people who were
currently accepted lUD were less likely to have discussed these issues with their
wives in the year preceding the survey than those currently not accepted.
Men in the action/maintenance stage were 2.0 times more likely to have low cons
for lUD use than those in the lower stage (OR=2.0; 95% confidence interval 1.1 to
3.7; P=0.02), and those in the action/maintenance stage were 3.7 times more likely
to report a high self-efficacy for lUD use than those in the lower stage (OR=3.7;
95% confidence interval 1.3 to 10.6; P=0.02).
127
O
O
O
O
ON
ON
vo
ON
in
vd en
CN
O
d
^o
en
o
CN
O
'C
es
u
es
9i
U
a
o
es
t
9i
en
0.9-
a
o
U
v
S
-d
d
C
2.0-
0.2-
-d
o
0.3-
9i
"d
o
OC}
OC}
en
r;
en
vo
en
I/)
ON
es
l>
en
00
OO
-d
d
ON
r00
OS
o t> R ^
m
TT
r- '1 o en
"es
CN
O
o
o
b
oo
rb
o
o
i>
as
o
m
in
en
o
in
in
r~;
en
en1
oo
CN
O
9i
a
d
es
ON
a
o
vo
t-^
en
en
en
9J
CN
CN
ri
CN
6s
iS
u
i-H
'-'
in
CN
'1
en
CN
*>
es
in
en
-d
in
<d
en
<z>
00
OCD
en
ON
en
^"
CN
CN
in
en
en
o en
Ov <^
g r^
i n ^ O ^ '-I
^ r^ CN o en
o
o
o
o
s
l4-l
o
CO
o
o
u
CO
CO
o
o
o
o
.d
.1t
<->
o
ex
'o
M
G
O
CO
CO
&0|
O
X ^
OH
00
CN
factors: socio-demographic
This would have been because they learned how to carry out "family
planning" from others in the community and may not have considered it as a
method without knovnng the exact of the method.
In the bivariate analysis, the significant differences between three stages of men's
readiness to accept lUD were revealed for socio-demographic variables (having a
son and wife with abortion history); confraceptive knowledge (spontaneous recall of
modem and fraditional confraceptive methods); communication with wives on
family planning issues; and social cognitive factors (pros and cons for lUD use and
self-efficacy for confraception in general and lUD use).
Two logistic regression models were computed. The first model compared men in
the precontemplation stage witii tiiose in the contemplation/preparation stage; the
second model compared men in the action/maintenance stage with those in the
contemplation/preparation stage. In the logistic estimations, six variables were
found as significant predictors for stage of men's readiness to accept lUD use. The
129
to
men
in
the
precontemplation
stage,
men
in
the
contemplation/preparation stage were about 5 times more likely to report high selfefficacy for lUD use. Those men were more likely to report higher confidence in
their ability to convince their wives to use lUDs than those m the precontemplation
stage. Compared to men in the contemplation/preparation, men in the
action/maintenance also were about 2 times more likely to report low cons for lUD
use and 3 times were more Ukely to report high self-efficacy for lUD use.
In both logistic regression models, self-efficacy for lUD use was the most important
predictor for men's readiness to accept the lUD (P<0.05). In the second model, the
finding may be explained by the fact that men whose wives were afready using
lUDs (action/maintenance stage) had more experience and therefore, more
confidence in convincing wives to use (and maintain use of) lUDs than did those
who had not (contemplation/preparation stage).
The findings also show that decisional balance for lUD use followed the expected
pattem for pros and cons in TTM studies. Pros for lUD use increased from the
precontemplation to action/maintenance stage, and the reverse pattem was reported
for cons for lUD use. The crossover was observed at the preparation stage. In
contrast, the pros and cons for contraception in general did not significantly increase
or decrease by stage of readiness for lUD use. Moreover, self-efficacy for
confraception in general decreased from the precontemplation to action/maintenance
stage. Thus, this confirms thefindingfromthe pilot study that specific measures for
lUD use were more sensitive in measuring men's readiness to accept the lUD.
Men in the action/maintenance stage were almost 3timesmore Ukely to have wives
who had not imdergone abortion. In other words, abortion has an inverse
relationship with lUD adoption. Men who accepted the lUD for confraception were
less likely to have wives who had undergone abortion than those who had not.
130
hi the study, men in the action/maintenance stage were ahnost 2 times more likely
to have a son in the family. After fulfilling their desfre for a son, men were more
likely to accept lUDs for confraception.
Men m the action/maintenance stage were more likely not to recall fraditional
method. They were currentiy using lUDs, tiierefore, tiiey did not need to use
fraditional methods and less likely to recall these methods. In comparison, those in
the precontemplation stage were more likely to recallfraditionalmetiiodstiianthose
in the contemplation/preparation stage. The findings suggest that those in the
precontemplation stage were more likely to use fraditional methods than those
thinking about lUD use.
Men in the action/maintenance stage were more likely to have low communication
with wives on family planning issues than those in a lower stage. lUD use does not
require much ongoing effort; lUD users may not need to discuss confraceptive
issues to the degree that users of condoms or the pill might need to. In confrast,
those thinking about adopting the lUD method may need more extensive discussion;
high communication was observed among those in the contemplation/preparation
stage.
A worthwhile intervention implied by the present findings might involve devoting
resources to men not considering lUD use (the precontemplation group, 29.5
percent of the study's sample), that is, to those least convinced of the need for lUD
confraception.
confidence in their ability to use lUDs and were more likely to use fraditional
methods. It might be beneficial to improve their self-efficacy for lUD use and to
provide them with information on the low effectiveness offraditionalmethods.
Another implication of the findings is the value of providing men at the
contemplation/preparation stage with information about lUD use. This group was
small but not an insignificant percentage of the study's sample (10.6 percent). Men
in this group reported much higher cons for lUD and lower self-efficacy than those
131
at the action/maintenance stage. Therefore, the sfrategy would be to increase selfefficacy and reduce perceived costs of lUD use.
The cross-sectional study was carried out with the aim of finding the modifiable
factors of stages of men's readiness to accept lUD use. The study shows the
identified predictors helpful for developing sfrategies for a stage-targeted
intervention promoting men's readiness to accept lUDs for contraception.
132
133
The theory (TTM) states that stage-targeted mterventions are needed to modify
factors. The intervention, m the form of relevant mformation promotmg lUD use,
should reduce misconception about cons; mcrease self-efficacy; discourage the
preference for sons; emphasize risk of abortion; increase confraceptive knowledge;
and encourage the couple to communicate about family planning.
The three stages of change identified for this particular study (Chapter 5) were
'precontemplation' (non-acceptance of the lUD and with no intention to change in
the next sbc months); 'contemplation/preparation' (intention to change in the next
30 days to 6 months); and 'action/maintenance' (acceptance of the the lUD for
confraception and having and lUD for six months or longer).
The stage-targeted intervention was designed to assist the movement of men
tiirough the stages of change for lUD use, that is, from the precontemplation stage
to the contemplation/preparation stage and into the action/maintenance stage. The
intervention program was developed based on concepts of TTM (Weinstein 1998;
VeUcer et al. 2000) using predictors of stages of men's motivational readiness to
accept the lUD for contraception.
TTM intervention studies have shown the use of stage-targeted letters and
counselling produced changes in different public health programs (VeUcer et al.
2000; PefrocelU 2002). For example, stage-targeted letters were found usefiil in
achieving higher scores on knowledge and behaviour associated with food choices
among adults who received the intervention (Davis et al. 2000); and counselling
was successful in promoting physical activity among sedentary patients in a health
care setting (King et al. 1998).
The intervention program was designed to employ two devices: stage-targeted
letters and interpersonal counselling. The interpersonal counselling was included to
ensure follow-up contact with participants and to ensure that all questions or
concems that arose during the intervention process were responded to.
TTM states that people in the precontemplation stage lack basic information of
behaviour change, a lack that should be addressed (Weinstein 1998). In comparison,
134
with effectiveness up to 99%, and with minimum side effects. That means, only one
person can get pregnant among J 00 women using RJD. The RJD version Tcu-380A
is available in An Hai District at this time and is effective for 10 years. The RJD is
very convenient to use; your wife needs to get it inserted ONCE only, and you do
not need to worry about unwanted pregnancies any more It is provided free of
charge at any district health centre and commune health centre, and can be
removed any time upon request. RJD can be inserted 42 days after a birth or an
abortion. It helps to prevent ectopic pregnancy when it is in place, and is safe for
women who have low exposure to STDs or HIV/AIDS like women in the An Hong
and Quoc Tuan communes ".
Each message was developed in a similar manner based on the concepts explained
earlier. The sample letter has been printed in Viettiamese and the English franslation
of letter is provided in Appendix F.
A quasi-experimental design was selected for the intervention. The most rigorous
evaluation design is the tme experiment, in which individuals are randomly
assigned to either group (intervention and comparison). In many field situations,
however, implementing a true experiment may be impractical (Fisher et al. 1985).
In a quasi-experimental design, the participants were not randomly assigned to
certain group (intervention or confrol). In this study, the village was chosen as the
primary unit for intervention. This design is called a moderate quasi-experimental
design and it has the advantages of fransparency, that is, it is known what setting
(village) will receive the intervention and what setting will not receive the
intervention (Gliner and Morgan 2000). The design was the most appropriate for an
intervention that used messages to promote behaviour change. There was a high
137
probability that men would pass on the messages (intervention) to others assigned to
the confrol groups, especially if they were in the same village.
6.3.2. Intervention sites
Sbc villages were chosen for intervention from the 12 villages identified m the
baseline survey. In order to reduce contamination during the intervention process,
three viUages were purposively selected from each commune after careful review of
their location. The intervention villages were selected for thefr separateness from
the confrol villages. The main road was chosen as the main criterion for separation.
Three villages on one side of the road in each commune were selected for the
control group, and the others for intervention. The distance between those groups
was 2-3 km. The distance between the two communes was 8 km. The maps of these
villages and communes are shown in Appendix A.
6.3.3. Intervention activities
The participants in the intervention villages received letters corresponding to their
stage of change for lUD use. They were encouraged to ask thefr local health
workers any question relating to the letter. Participants in the confrol villages
received neither letter nor interpersonal counselling. In both intervention and
confrol villages, the local health workers continued their routine activities.
Two rounds of intervention were carried out over a period of six months. The first
round was conducted in November 2001. This involved a visit to each participant
from a local health worker who used a staging algorithm to assess the men's current
stage of change for lUD use. Each participant was provided with a letter
corresponding to his identified stage of change for lUD use. The local health
workers asked men to read through the letter, to record any questions if provoked
and to discuss the letter with their wives and others. After one week, the local
health workers visited the men again, answered any questions and collected
information on the men's perception regarding the stage-targeted letter.
138
The second round of intervention took place after three montiis, m Febmary 2002.
The staging algorithm was used to assess the men's current stage of change for lUD
use. Each participant received a letter correspondmg to his new stage of change for
lUD use. The main purpose of the second round was to remforce any commitment
to behaviour change, to encourage communication and adoption of the lUD. Men
were asked again to read the letter, to record questions and were encouraged to
discuss family planning and lUD use with wives and otiiers. Similar to the first
round, the local health worker visited again one week later to answer questions and
to assess the men's perception regarding the stage-targeted letter.
6.3.4. Training and supervision offieldworkers
The local health worker in each intervention village was assigned to carry out the
mtervention activities. Before each intervention round, those health workers
attended two days intensive fraining conducted by the researcher. They received the
protocol on how to carry out the intervention. Information on all activities and the
information that they were required to provide during each intervention round was
included in the protocol. The intervention protocol is in Appendix G. During
fraining, the protocol was discussed at length and the researcher tested performance
by engaging in a role-play with the health workers' supervisor. In addition, the
researcher and supervisor were available by telephone for consultation, problem
solving and encouragement during thefieldworkperiod.
Two district health workers in charge of the family planning program supervised the
local health workers during the intervention period. They randomly checked 20
percent of all intervention households, to see if the intervention was carried out
correctly. Those health workers who had not correctly performed the tasks were
requested to repeat them.
During the six months of intervention, all events relating to family planning from
both intervention and confrol villages were recorded in monitoring forms and
submitted to the researcher (see the monitoring form in Appendix G).
139
140
Percentage (%)
Effectiveness of lUD
99.7
91.4
Pregnancy susceptibility
91.4
Abortion risk
97.9
Communication on lUD
67.1
Adoption of lUD
73.9
Others
2.4
The reaction to the appearance and content of the letter was measured by a Likerttype scale from 1 to 7 (not attractive/interesting to very atfractive/interesting).
Table 6. 2. Men's perception of the stage-targeted letters
Evaluation items (N=337)
Mean
SD
6.3
0.6
6.4
0.6
6.4
0.6
Men found the stage-targeted letter atfractive in its appearance (mean = 6.3; SD =
0.6). The letter's content was interesting (mean = 6.4; SD = 0.6) and useful (mean =
6.4; SD=0.6). The findings suggest that the letters were well designed and contained
useful information, atfracting the attention of participants (Table 6.2).
141
CA
50 o>
ra
[IlEasy
40-
Very easy
D Neither easy nor difficult
30 J
20 -
' ^-^^^^l
10 ^^^H^^^^l
_i
Understanding
Figure 4 presents the men's perception of thefr level of understanding after reading
the letter. The level of understanding was categorized into 5 levels, from very
difficult to very easy. Figure 4 shows that information was not difficult. None of
participants indicated the information was difficult or very difficult to understand.
More than two-thirds (67.7 percent) indicated the information was easy; less than
one-third (29.4 percent) perceived it was easy, and a small proportion (3.0 percent)
reported it was neither difficult nor easy.
Figure 5. Perceived relevance
ft
100.0 -
B
V
50.0^
40.0-
"
30.0 -
Low
D Moderate
DHigh
RELEVANCE
142
Figure 5 presents results on how relevant the letter was to the men on lUD use. The
question used a rating scale from 1 to 7 (not relevant to very relevant) and tiie level
of relevance was categorized mto low, moderate and high. A majority (94.4 percent)
of men perceived the letter was highly relevant to tiiem. Very few stated its level of
relevance was low or moderate (about 3 percent).
Figure 6 presents the effects of the intervention letters on whetiier or not tiie letters
were showed to or discussed with others (wives, family planning collaborators,
health workers, fiiends, family members). Less than two-thfrds (62.3 percent) of
men showed the letters to friends/family members, and about half (49.9 percent)
showed it to health workers/family planning collaborators. Ahnost all participants
(99.4 percent) discussed the letters with wives and less than half (43.6 percent)
discussed them with others.
Figure 6. Communication effects
nvvi\s
n Other people
Health worl<ers
Friend/Relatives
o
sz
20
40
60
80
100
120
Percentages
143
120"
100
an
re
c
60
0)
O)
<0
0
iSucceded in convincing
I Total men in each group
k.
Q.
40
20
To get lUD
To continue use
lUD
6.5. Conclusion
The stage-targeted intervention was designed to promote men's readiness to accept
lUDs for confraception and it was comprised of stage-targeted letters and
interpersonal counselling. The stage-targeted letters contained items of information
found to be the significant predictors by the baseline survey. Three different letters
were designed for people in the three different stages: precontemplation;
contemplation/preparation and action/maintenance.
The intervention was carried out in two rounds during a six-month period. The local
health worker in each village was assigned to conduct the intervention activity: to
deliver the stage-targeted letters to participants, and answer thefr questions. District
health workers in charge of family planning and the researcher closely supervised
the intervention activities. Any question or problem that arose during
implementation was responded to by the researcher and/or supervisors. The overall
evaluation of the stage-targeted letter was every encouraging. The men generally
reported that the letter was well written, well presented and atfractive. Moreover, it
encouraged men to show the letters to, and discuss the contents with, thefr wives
and otiier people. The information was useful for them in convincing wives to adopt
lUDs or to continue to use lUDs.
144
Hypothesis 1. Men, who have received the stage-targeted mtervention program, are
more likely to report significantly higher spontaneous recall of modem
confraceptive methods than those who have not.
Hypothesis 2. Men, who have received the stage-targeted intervention program, are
more likely to report significantiy higher spontaneous recall of traditional
confraceptive methods than those who have not.
Hypothesis 3. Men, who have received the stage-targeted intervention program, are
more Ukely to report significantly higher communication with their wives on family
planning issues than those who have not.
Hypothesis 4. Men, who have received the stage-targeted intervention program, are
more likely to report significantiy higher communication with others on family
planning issues than those who have not
Hypothesis 5. Men, who have received the stage-targeted program, are more likely
to report significant progress through the stage of change for lUD use than those
who have not.
Hypothesis 6. Men, who have received the stage-targeted program, are more likely
to report significantly higher pros for confraception in general than those who have
not.
Hypothesis 7. Men, who have received the stage-targeted program, are more likely
to report significantly higher pros for lUD use than those who have not.
Hypothesis 8. Men, who have received the stage-targeted program, are more likely
to report significantly lower cons for confraception in general than those who have
not.
Hypothesis 9. Men, who have received the stage-targeted program, are more likely
to report significantly lower cons for lUD use than those who have not.
146
Hypothesis 10. Men, who have received the stage-targeted program, are more Ukely
to report significantiy higher self-efficacy for confraception in general than those
who have not.
Hypothesis 11. Men, who have received the stage-targeted program, are more likely
to report significantiy higher self-efficacy for lUD use than those who have not.
7.2. Data collection
Face-to-face interviews were carried out to collect data using a cross-sectional
survey. All participants in the baseline survey in both intervention and confrol
villages were identified and contacted for interview. The researcher conducted twoday training for interviewers, the same local health workers involved in the baseline
study. The fraining for interviewers and the supervision procedure were similar to
that of the baseline study and is described in Chapter 3 (section 3.4).
7.3. Measures
Measures, similar to the baseline study measures, were used in the questionnafres of
the posttest study (Appendix K). The measures were organised into the following
sections:
1. socio-demographic information;
2. knowledge on family planning;
3. commimication on family planning issues;
4. staging algorithm for lUD use;
5. pros and cons and self-efficacy for both confraception in general and lUD use.
The measures are described in Chapter 5 (baseline survey). The reliability and
means and SD of each consttnct are provided in Table 7.1. The intemal consistency
reUability of each constmct was estimated by Cronbach's a coefficient, which
varied from 0.79 to 0.9 for all scales. The results showed good mtemal consistency
reUability.
147
Cronbach's
Means
coefficient
Standard
Deviation
0.9
17.4
1.9
0.79
5.6
1.2
0.87
12.5
1.8
0.9
10.4
4.0
general
0.84
19.2
3.0
0.89
14.9
3.6
148
differences of scores between the intervention group and the confrol group at the
posttest to enable the analysis to the baseline differences.
7.5. Results
The results are presented in four sections. Ffrst section presents the resuhs of tests
of equivalence: between the posttest group and lost to follow-up group and between
the intervention group and the confrol group at the baseline study. Second section
illusfrates the change in confraceptive knowledge and communication from baseline
to posttest. Third section reports the change in stages of readiness to accept lUD use
from baseline to posttest. Finally, the fourth section presents the change in pros,
cons and self-efficacy for contraception in general and lUD use from baseline to
posttest.
7.5.1. Test of equivalence
Of the 651 participants who completed the baseline interview, 354 were in
intervention villages and 297 were in confrol villages. Only a total of 610 men could
be followed-up at the posttest interview, with 336 men in the intervention group and
274 in the confrol group. A total of 41 people were lost to follow-up at the posttest
survey; 18 persons were in the intervention group and 23 persons were in the
confrol group. The follow-up rate at posttest was 95 percent for the intervention
groups and 92 percent for the confrol groups.
Test of equivalence between posttest and lost to follow-up group
The information on the difference between those included and those lost to followup in the posttest by socio-demographic factors, confraceptive knowledge,
communication, stages of change for lUD use and cognitive constmcts are
presented in Appendix J.
There was no statistically significant difference (P>0.05) between participants in the
posttest and those lost to follow-up by age, education, occupation, ethnicity,
reUgion, abortion history, last birth wanted, knowledge on confraceptive methods
150
151
Study group
Intervention
Confrol
(N=274)
(N= 336)
P value
Age groups
19-24
2.1
2.6
25-29
30-34
15.5
13.1
20.8
25.2
35-39
40-44
35.1
35.4
24.4
21.2
45-49
2.1
2.5
Primary
Lower secondary
8.9
65.5
5.9
55.5
Upper secondary
22.9
2.7
27.7
10.9
11.9
16.7
21.9
17.2
67.3
4.1
54.7
6.2
25.6
32.1
51.2
23.2
55.5
12.4
No
22.3
28.5
Yes
77.7
71.5
75.0
70.1
25.0
29.9
No
70.2
78.8
Yes
29.8
21.2
No
28.9
25.5
Yes
71.1
74.5
0.7
Education
Higher education
Occupation
Government
Private
Agricultural
Otiier
Number of children
One
Two
Three and more
000
0.00
0.00
Having a son
0.09
Desired children
No
Yes
Wife's history of abortion
0.17
0.02
152
153
Study group
P value
Intervention
Confrol
(N== 336)
(N=274)
93.2
94.9
6.8
5.1
Recall
39.3
36.9
No recall
60.7
63.1
Low
97.9
93.1
High
2.1
6.9
Low
80.7
76.3
High
19.3
23.7
Precontemplation
28.6
29.2
Contemplation/Preparation
11.6
9.9
Action/Maintenance
59.8
60.9
No recall
0.3
0.5
0.003
0.1
0.78
normaUty, linearity, univariate and multivariate outUers, homogeneity of variancecovariance mattices, and multicollinearity (Pallant 2001). No serious violation of
assumption for the MANOVA was detected. A statistically significant difference
between groups on the combined dependent variables (pros and cons for lUD) (F
(1,609) = 6.341; P<0.05) was found. The results of one-way ANOVA revealed tiiat
the two groups were significantly different on pros and cons for lUDs. Men in
intervention group had higher meais of pros for lUDs (51.3 vs. 48.4; P<0.05) and
lower cons for lUDs (48.4 vs. 51.4; P<0.05) than those in confrol group.
Table 7.4. Differences (Mean, SD) of TTM constructs at baseline, by study
group
Characteristics
Study group
F ratio
Intervention
Confrol
(N=336)
(N=274)
P value
50.7
49.2
SD
10.1
9.9
Mean
49.5
50.5
SD
10.4
8.8
Mean
51.3
48.4
SD
9.2
10.5
Mean
48.4
51.4
SD
9.3
10.5
Mean
49.9
50.4
SD
10.0
9.8
Mean
50.1
49.9
SD
10.0
10.0
3.304
0.07
1.756
0.18
13.133
0.00
14.090
0.00
0.386
0.53
0.037
0.84
Self-efficacy for
confraception
155
MANOVA tests were carried out for self-efficacy for confraception in general and
lUD use; no significant differences were detected.
7.5.2. Change in contraceptive knowledge and communication level
Four outcome measures that addressed men's readiness to accept tiie lUD for
confraception during the follow-up period were assessed: confraceptive knowledge
(spontaneous recall of modem and traditional confraceptive methods) and
communication with wives and others on family planning issues. The Chi-square
tests were performed to identify the differences between the two groups at tiie
posttest study. The results are presented in Table 7.5
Table 7. 5. Percentage distribution of contraceptive knowledge and
communication at posttest survey, by study group
Outcome variables
Intervention
Control
Chi-
(N=336)
(N=274)
square
P value
Spontaneous recall of
modem methods
99.7
98.2
3.61
0.06
Spontaneous recall of
fraditional methods
71.7
36.9
74.47
0.00
63.7
36.3
96.0
4.0
92.34
0.00
54.5
45.5
98.9
1.1
156.60
0.00
Results of McNemar test identifying tiie change between baseline and posttest on
spontaneous recaU of modem and fraditional methods and communication with
wives and others on family planning, for each study group, are presented in Table
7.6.
156
Intervention
P value
(N=336)
Spontaneous recall of
modem methods
Baseline
Posttest
93.2
99.7
Spontaneous recall of
fraditional methods
Baseline
Posttest
39.3
71.7
Control
P value
(N=274)
0.00
94.9
98.2
0.05
0.00
36.9
36.9
1.00
19.3
36.3
0.00
23.7
4.0
0.00
2.1
45.5
0.00
6.9
1.1
0.00
157
96
c
S 94
Intervention
Control
a>
a.
92
90
88
Baseline
Posttest
Intervention
g 40
Control
S. 30
20 ^
10
0
Baseline
Posttest
158
- Intervention
- Control
I 20
u
15
CL
10
5
0
Baseline
Posttest
159
Thus, the resuUs support the thfrd hypothesis: men who had received intervention
reported significantiy higher communication with wives on family planning issues
than those who had not.
The resuhs in Table 7.6 show the differences between baseline and posttest for the
intervention group as well as for the control group. The proportion of men who had
high communication with others in the intervention group significantly increased
from 2.1 to 45.5 (P<0.05). In confrast, the proportion of men who had high
communication with others in the confrol group significantly decreased from 6.9 to
1.1 percent (P<0.05) (Figure 11).
Intervention
S?o
he
Control
S20
^ 5
10
5
0
Baseline
Posttest
Thus, the resuhs support the fourth hypothesis: men who had received the
intervention reported significantly higher communication with others on family
planning issues than tiiose who had not.
160
Intervention
Control
Chi-
use
(N=336)
(N=274)
square
Precontemplation
20.2
33.9
5.4
5.8
74.4
60.2
Contemplation/Preparation
Action/Maintenance
226.432
P value
0.00
Intervention
P value
Control
P value
(N=274)
(N=336)
Precontemplation
Baseline
Posttest
28.6
20.2
29.6
33.9
Contemplation/Preparation
Baseline
Posttest
11.6
5.4
9.9
5.8
Action/Maintenance
Baseline
Posttest
59.8
74.4
60.9
60.2
0.00
0.54
161
The McNemar test showed a significant change between baseline and posttest on
SOC for the intervention group (P<0.05), but not for the confrol group (P>0.05)
(Table 7.8). In the intervention group, the proportion of men in the
precontemplation stage decreased from 28.6 percent at baseline to 20.2 percent at
posttest, and the proportion of those in the contemplation/preparation stage
decreased from 11.6 to 5.4 percent. As expected, the proportion of men in the
action/maintenance stage significantly rose from 59.8 to 74.4 at the posttest (Figure
12). In confrast, in the confrol group, the proportion of men in the precontemplation
stage increased from 29.2 percent at baseline to 33.9 percent at posttest; the
proportion in the contemplation/preparation stage decreased from 10.2 percent at
baseline to 5.8 percent at posttest; and the proportion of men in the
action/maintenance stage remained almost the same (60.6 and 60.2 percent) (Figure
12).
Figure 12. Change in SOC for lUD use
80
70
60
50
40
30
20
10
0
w
O)
re
*.>
c
u
Q.
(3 Precontemplation
Contemplation/Preparati
on
D Action/Maintenance
l
d.cS
/
/ "
0"
^.c^'
.^
<6'
<?^'^
hi summary, the resuhs support the fifth hypothesis: men who had received tiie
intervention were more likely to move to tiie action/maintenance stage and accept
lUD use for confraception than those who had not.
162
7.5.4. Change m decisional balance and self-efficacy for contraception and lUD
use
Six outcome measures that addressed men's readmess to accept lUDs for
confraception during the follow-up period were assessed: pros and cons for
confraception in general and lUD use, and self-efficacy for confraception m general
and lUD use. A MANOVA was performed to check tiie differences of decisional
balance between the intervention group and the confrol group at posttest.
Assumptions were tested to check for normality, linearity, univariate and
multivariate
outiiers,
homogeneity
of variance-covariance
matrices, and
multicolUnearity (Pallant 2001), and no serious violation were detected. There were
statistically significant differences between groups on the combined dependent
variables (cons for confraception in general, pros and cons for lUD use) (F (1,609)
= 107.258; P<0.05). No significant difference between two groups was found for
pros for confraception in general at posttest. Therefore, no intervention effect was
found for pros for confraception in general, and the mixed between-within subject
ANOVA was carried out only for cons for confraception, and pros and cons for lUD
use.
Table 7. 9. Baseline and posttest means on selected cognitive variables
Cognitive variables
Baseline
Posttest
(Mean, SD)
(Mean, SD)
336
274
50.7(10.1)
49.2 (9.9)
49.9(10.2)
50.1(9.7)
336
274
49.5 (10.4)
50.5 (8.8)
46.3 (10.9)
54.6 (6.4)
336
274
49.9 (10.0)
50.4 (9.8)
51.1(10.8)
48.7 (8.8)
336
274
50.1 (10.0)
49.9 (10.0)
54.6 (6.3)
48.4 (6.8)
163
IntervAention
Control
46 44 42
Baseline
Posttest
164
Thus, the resuhs support the eightii hypotiiesis: men who had received the
intervention, were significantly more likely to report lower cons for confraception in
general than those who had not received the intervention.
Change in self-efficacy for contraception in general
The resuhs of a mixed ANOVA revealed the main effect within subject (time) was
not significant for self-efficacy for confraception in general. Thus, no significant
change from baseline to posttest across the two study groups was detected (F (1,
609)=0.215; P>0.05). Similarly, the main effect between subjects (study groups)
was not significant (F(l,609)=2.411; P>0.05); therefore, there was no significant
difference among study groups on self-efficacy for confraception in general across
the baseline and posttest period.
There was statistically significant interaction between group and time (time x
group) (F(l, 609)=6.831; P<0.05), therefore, one-way ANOVA test was conducted
to identify the difference between the groups at posttest. The results, as expected,
showed that there was statistically significant higher self-efficacy for contraception
in general for the intervention group than the confrol group (51.1 vs. 48.7; F (1,
609)=8.397; P<0.05) (Figure 14; Table 7.9)
Figure 14. Change in self-efficacy for contraception in general
51.5
51
50.5
50
- Intervention
8 49.5-
- Control
(0
49
48.54847.5
Baseline
Posttest
The repeated ANOVA was performed to test the simple effect of test time within
each study group. However, there was no significant increase of self-efficacy from
165
baseline to posttest for the intervention group (49.9 vs. 51.1; F(l, 335)= 2.499;
P>0.05), however, there was a significant decrease from baseline to posttest for tiie
confrol group (50.4 vs. 48.7; F(l, 273)=4.465; P<0.05) (Figure 14; Table 7.9).
Thus, the resuhs showed no support for the tenth hypothesis: men who had received
the mtervention did not report significantly higher self efficacy for confraception m
general than those who had not received the intervention. Although the posttest
score of tiie intervention group was significantly higher than tiie confrol group, tiiere
was no significant mcrease from baselme to posttest on self-efficacy for
confraception in general for the intervention group.
Change in self-efficacy for RJD use
Figure 15. Change in self-efficacy for lUD use
56
55
54 -\
53
52
S 51
8 50
- 49^
48 4746
45
Intervention
Control
Baseline
Posttest
The results of a mixed ANOVA showed the statistically significant main effect
within subject (time) for self-efficacy for lUD use; therefore, there was a significant
change from baseline to posttest across the two study groups (F (1, 608)=14.887;
P<0.05). The main effect between subjects (study groups) was also significant
(F(l,608)=32.801; P<0.05). Hence, there was significant difference among study
groups across the baseline and posttest period.
Furthermore, there was statistically significant interaction between group and time
(time x group) (F(l, 608)=56.585; P<0.05), therefore, ANOVA test was performed
166
to identify the difference between the groups at the posttest. Results, as expected,
showed that there was significantly higher self efficacy for lUD use reported for tiie
mtervention group than the confrol group (54.6 vs. 48.4; F (1, 608)=135.611;
P<0.05) (Figure 15; Table 7.9).
Figure 15 shows that the frends of self-efficacy for lUD use for the two groups from
basehne to posttest were not sunilar. The repeated ANOVA was performed to test
the simple effect of test time within each study group. As expected, there was
significait increase from baseUne to posttest for the intervention group (50.1 vs.
54.6; F(l, 335)= 5.866; P<0.05), and the significant decrease from baselme to
posttest for the confrol group ( 49.9 vs. 48.4; F(l, 273)=74.433; P<0.05) (Figure
15).
Thus, the results showed support for eleventh hypothesis: men who had received the
intervention, were significantly more likely to report higher self-efficacy for lUD
use than those who had not received the intervention.
Change in pros and cons for RJD use
Table 7.10. Baseline and posttest means on pros and cons for lUD use
Group
Baseline
Posttest
Adjusted
(Mean, SD)
(Mean, SD)
posttest
(Mean, SE)
336
274
51.3 (9.2)
48.4(10.5)
53.0(8.1)
46.3 (10.8)
52.7 (0.50)
46.6 (0.55)
336
274
48.4 (9.3)
51.4(10.5)
44.8 (6.6)
56.3 (9.8)
45.0 (0.44)
56.0 (0.48)
baseline scores, there was significant higher pros for lUD in the intervention group
than the confrol group on posttest scores (52.7 vs. 46.6; F(l, 608)=65.965; P<0.05)
(Figure 16).
Figure 16. Change in pros for lUD use
54-1
53
52
51
50
49-1
47-1
46
4544 43
o 48
u
Intervention
Control
Baseline
Adjusted posttest
Similarly, an ANCOVA test carried out to identify the intervention effect on cons
for lUD use, the independent variable was the study group (intervention and
confrol), and the dependent variable was scores of cons for lUD use at the posttest.
Participants' scores on the baseline were used as the covariate in this analysis (as
there was significant difference between groups at baseline scores on cons for lUD
use). After adjusting for baseline scores, there was a significantly lower cons for
lUD use in the intervention group than the confrol group on posttest scores (45.0 vs.
56.0; F(l, 608)=280.739; P<0.05) (Figure 17; Table 7.10).
168
60
50
40
- Intervention
O30
- Control
10
0
Baseline
Adjusted posttest
Thus, the resuhs showed support for the seventh and mnth hypotheses: men who
had received the intervention were significantiy more likely to report higher pros for
lUD use and lower cons for lUD use than those who had not received tiie
intervention.
7.6. Summary of findings
This quasi-experimental design study demonsfrated the feasibiUty of delivering a
stage-targeted intervention to married men in rural Vietnam to promote men's
readiness to accept lUDs for confraception. The study followed-up 610 of the 651
participants to provide a 92.0 and 95.0 percent follow-up rate for the confrol group
and the intervention group, respectively. Those lost to follow-up were people who
were more Ukely to have one child, less likely to have a son in the family, less likely
to desire more children, more likely to have high communication with wives on
family planning issues and also more likely to report higher cons for lUDs.
Table 7.11. Summary of findings on hypotheses testing
Hypothesis
Results
169
Men, who have received the stage-targeted intervention program, are Supported
more likely to report significantly higher communication with thefr
wife on family planning issues than those who have not.
Men, who have received the stage-targeted intervention program, are Supported
more likely to report significantly higher communication with others
on family planning issues than those who have not.
Men, who have received the stage-targeted program, are more likely Supported
to report significant progress through the stage of change for lUD
use than those who have not.
Men, who have received the stage-targeted program, are more likely Not
to report significantly higher pros for contraception in general than supported
those who have not.
Men, who have received the stage-targeted program, are more likely Supported
to report significantly higher pros for lUD use than those who have
not.
Men, who have received the stage-targeted program, are more likely Supported
to report significantly lower cons for contraception in general than
those who have not.
Men, who have received the stage-targeted program, are more likely Supported
to report significantly lower cons for lUD use than tiiose who have
not.
Men, who have received the stage-targeted program, are more likely Not
to report significantiy higher self efficacy for confraception in supported
general than those who have not.
Men, who have received tiie stage-targeted program, are more likely Supported
to report significantly higher self efficacy for lUD use tiian tiiose
who have not.
At baselme, some differences between intervention and confrol groups were seen by
education level, occupation, parity, communication witii otiiers, pros for
confraception in general and pros and cons for lUDs. The differences for pros and
cons for lUDs were confroUed for in the posttest analyses.
The intervention improved the recall of modem confraceptive metiiods in the
intervention group; however, tiiere was no significant difference between groups at
posttest. hi confrast, tiie intervention significantly increased recall of fraditional
methods of men in the intervention group.
170
The intervention facilitated higher communication with otiiers and wives on family
planning issues. One purpose of the message in the stage-targeted letter delivered to
participants, was to encourage discussion between men and wives and others on
family plannmg (as well as on lUD use). This message produced a significant
unpact on communication. While at baselme the percentage of men who had high
commuiucation with wives as well as with others was lower for the mtervention
group, at posttest, the intervention group reported significantly higher
communication than reported by the confrol group.
This posttest study provides sfrong support for the TTM-mspired stage-targeted
intervention to promote lUD acceptance. Intervention effects were revealed for
lowering cons for confraception in general and lUD use in particular and for
increasing pros and self-efficacy for lUD use. However, no intervention effect was
found for increasing pros and self-efficacy for confraception. As hypothesized, men
in the intervention group reported significantly lower cons for confraception in
general and RJD use and significantly higher pros for lUD use at posttest compared
to baseline, even after confroUing for baseline differences (pros and cons for lUD
use). The men in the intervention group also reported significantly lower cons for
confraception in general and higher self-efficacy for lUD use than those in the
confrol group at posttest.
An important outcome of the intervention was the movement through stages of
change for lUD use. It was hypothesized that men who received the intervention
were significantly more likely to progress from the lower to higher stages of change
for lUD use. The proportion of men in the action/maintenance group in the
intervention group showed a significant increase from 59.8 to 74.4 percent at
posttest, while there was no change for the confrol group.
In conclusion, findings sfrongly support stage-targeted intervention using a TTM
framework in promoting men's readiness to accept the lUD for confraception.
171
CHAPTER 8: DISCUSSION
This chapter is organised into five sections. The ffrst section provides a brief review
of the main feattires of the study. The second presents tiie keyfindmgsof tiie sttidy
m relation to the wider literature, givmg particular attention to methodological
issues and limitations of the study. In the final section, recommendations for fiirther
research and family planning services in Vietnam are proposed.
8.1. Overview
Men are significantiy involved in family planning decision-making processes
(Salway 1994; Biddlecom et al. 1996; Bankole and Smgh 1998). The husband's
approval is considered by many researchers to be the most important predictor of
confraceptive use (Joesoef et al. 1988; Lasee and Becker 1997; Kamal 2000). The
overaU goal of the study was to identify methods by which targeted health
behavioural change programs might be best developed and implemented to increase
acceptance of modem confraception in rural Vietnam.
Among all modem confraception, the lUD is the most commonly used method in
Vietnam and about two-thirds of people report that they ever-used lUDs (NCPFP
1999). The lUD is a very effective reversible method for prevention of pregnancy
and it is suitable for couples who have at least one child and low exposure to
STD/HIV/AIDS (Treiman et al. 1995; TmsseU et al. 1995; Hicks 1998). Obtammg
an lUD is very convenient and it is provided free of charge in any commune health
cenfre in Vietnam. Based on these reasons, the lUD was selected as the
confraceptive method for the study.
The study had three research questions: (1) to identify measures of men's readiness
to accept the lUD for confraception in rural Viettiam; (2) to identify potential
modifiable factors to each stage of men's readiness to accept the lUD for
confraception; and (3) to test whether an intervention targeted at each stage of
readiness will resuh in an increase in men's acceptance of the lUD for
confraception.
172
The sttidy used a quasi-experimental design. Two rural communes. An Hong and
Quoc Tuan, in An Hai Disttict, Hai Phong Provmce, were mvolved m tiie sttidy.
Two rounds of mtervention were carried out in six montiis of follow-up.
Participants in six villages in the mtervention group received stage-targeted letters
and mterpersonal counselling. Participants in the other six villages, constittiting tiie
confrol group, received neither letters nor counselling.
A review of literature on men's involvement in family planning revealed a number
of factors related to confraceptive behaviour. At the individual level, factors
included knowledge, approval, communication, reproductive preferences, and
socio-demographic factors (i.e. age, income, education and occupation). At the
program level, the factors included accessibiUty, availabiUty, costs, and quality of
services. At the community level, the factors mcluded family planning and
population policy, religion, social norms, etc. By selecting participants from tiie
same village, the study design confroUed for many of the program and community
level factors.
In Chapter 2, the main theories of behaviour change were reviewed and TTM was
chosen as the appropriate model for the study. Three constmcts from TTM were
used in the study: (1) stages of readiness to accept confraceptive methods and lUDs;
(2) decisional balance (pros and cons for confraception in general and lUD use); and
(3) self-efficacy for confraception in general and lUD use. Findings from a pilot
study showed that these constmcts were valid and reliable in examining men's
readiness to accept lUD for confraception in rural Vietnam (Chapter 4).
Confraceptive knowledge and communication on family planning issues were
identified for inclusion in the study as factors influencing men's acceptance of
confraceptive methods. In addition to socio-demographic variables, other important
factors that influenced men's acceptance of confraception in the study were fertility
intention, havmg a son, last birth intention, number of children and wife's abortion
history. The assumption of the study is that the more ready men are to accept lUDs
for confraception, the more likely they are to convince thefr wives to use, or to
continue to use, lUDs for confraception.
173
The baselme survey was carried out in March-April 2001 to obtafri mformation
from eligible men mtiieintervention and confrol villages. The collected mformation
covered: (1) socio-demographic characteristics; (2) confraceptive knowledge; (3)
communication with wives and with otiiers on family plannmg issues; (4) stages of
readiness to accept lUDs for confraception; and (5) decisional balance (pros and
cons) and self-efficacy for confraception and lUD use.
The results of the baseline study revealed six significant predictors in each stage of
men's readiness to accept lUDs for confraception. The predictors included TTM
constmcts (cons and self-efficacy for lUD); socio-demographic characteristics
(history of having an abortion and having a son); confraceptive knowledge
(spontaneous recall of fraditional methods) and communication on family planning
(couple communication). These findings were used to design an mtervention
targeting the stages of men's readiness to accept lUDs for confraception.
The intervention consisted of stage-targeted letters and interpersonal counselling.
The letters were developed using the principles of targeted health messages from
Kreuter et al (1999). The process assessment revealed that the stage-targeted letters
were well designed, with information relevant to the men, and had significant
impact in facilitating men's communication with wives and other people on family
planning (Chapter 6).
A posttest study was carried out after six months follow-up and showed support for
the stage-targeted intervention. The level of spontaneous recall of fraditional
methods rose significantly from 39.3 to 71.7 percent from baseline to posttest.
Likewise, the level of high communication with wives increased significantly from
19.3 to 36.3 percent, and high communication with others rose from 2.1 to 45.5
percent. The mean pros for lUD use in the intervention group rose from 51.3 to
52.7, while cons for lUD decreased from 48.4 to 44.8. The mean self-efficacy for
lUD increased from 50.1 to 54.6, while mean cons for confraception in general
decreased from 49.5 to 46.3. The intervention also had a significant impact on
moving people towards the action/maintenance stage of men's readiness to accept
lUDs for confraception. The proportion of men in the action/maintenance stage rose
from 59.8 to 74.4 percent (Chapter 7).
174
sample was 29 years old at that time, therefore, this policy mfluenced a majority of
participants and may have conttibuted to the small number of Uvmg children, and a
small family size.
Over two-thfrds of men did not want to have any more children. This findmg was
consistent witiifindingby Mai et al. (2001). In the past, before the market economy
was infroduced m 1986, each additional child was seen as a blessmg and an
economic asset. Children are no longer seen as economic assets, they are expensive
to bring up and to educate, and they require care when parents are at work. The
school fees, health care, and other expenditures related to child upbringing are a
heavy financial burden for parents. Therefore, a large proportion of men did not
wish to have any more children (Johansson et al. 1998b).
The desire to have no more children usually is positively associated with
contraceptive use. The use of confraception is higher among people who want to
stop childbearing (Bongaarts 1992). However, in the study, no significant
relationship between desire to have more children and stage of change for lUD use
was found. The reason may be related to the difference of analytical approach. In
other studies, contraceptive use was dichotomised (yes/no), while in this study the
use of the lUD was categorized by three stages of readiness to accept the lUD for
confraception.
Sttidies have shown that couples who have a son are more likely to practice
confraception than those who have not (Oyeka 1989; Stash 1996). Men who have
accepted lUDs for confraception are ahnost two tunes more likely to have a son m
the family than those have not. Although some recent sttidies have shown tiiat son
preference is less prominent in Vietiiam (Haughton 2000; Mai et al. 2001), resuhs
from tills sttidy stiU show support for the association of son preference and
confraceptive use.
One-fourth of men in this study's sample had wives who had undergone abortion.
The maximum number of abortions was four. Men who accepted the lUD for
confraception were more likely to have wives who had not had an abortion.
Abortion has an inverse relationship with lUD adoption and tiie use of lUD reduces
the risk of abortion. This finding is consistent with otiier sttidies. Gorbach et al.
(1998) examined confraception and abortion in two Vietnamese communes and
found the use of the lUD reduced the likelihood of subsequent abortions in these
communes by 70 percent. For rural people, the use offraditionalmetiiods increased
the likelihood of abortion by 1.66.
Although most of Viettiamese people officially are not clauned to be Buddhists,
they were brought up with Buddhists beliefs as Buddhist's followers (Belenger and
Hong 1998). According to the ethics of lay Buddhists in Vietnam, abortion is seen
as a severe sin. However, in societies where the confraceptive choice is limited,
abortion is likely to remain an important option for limiting family size (Kulszycki
etal. 1996).
8.2.2. Measures of men's readiness to accept lUD for contraception
A detailed review of the literature on behaviour change and confraception revealed
that perceived benefits, costs and self-efficacy are important social cognitive factors
in predicting confraceptive
use.
Confraceptive
177
TTM constructs
The cenfral constmct of TTM is stage of change. Men were categorized mto five
stages of readiness to accept lUDs for confraception by assessing thefr intention of
acceptmg lUDs for confraception. The stagmg algorithm adapted from Grimley and
Lee (1997) was used in a pilot study (Chapter 4). The appeal of tiie SOC is tiiat it
conceptuaUses change as a process of movement through a sequence of five stages
(Prochaska et al. 1992). ft was found that SOC measures based on TTM provide
sensitive assessment of readiness to accept the lUD for confraception (Chapter 4).
The SOC constmct has also been used for selecting appropriate intervention. By
locating a person in a specific stage of the change process, a targeted intervention
can be provided by customizing the message to meet his current needs, rather than
expecting his needs to match an action-oriented intervention program (VeUcer et al.
2000).
In the sample population, the baseline survey found above half in the
action/maintenance stage; less than one-third was in the precontemplation stage and
a very small proportion was in the contemplation/preparation stage. This
distribution is not consistent with other studies, where the highest proportion of
people was in the precontemplation or contemplation stages for condoms or the pill
(Grimley et al.l995; Galavotti et al.l995; Grimley and Lee 1997; Stark et al.l998).
The difference reflects the long history of the lUD as the dominant method of
confraception in Vietnam. Although other modem confraceptive methods have been
infroduced in Vietnam, tiie lUD is still the most common metiiod in the current
decade (NCPFP 1999).
The intermediate (dependent) consttiicts of TTM are decisional balance and self
efficacy. These measures are sensitive to progress through all stages (VeUcer et al.
2000). In this sttidy, the decisional balance constiuct examined tiie perceived
payoff for the men for adopting and continumg lUD use (i.e. decisional balance or
the pros and cons of accepting confraception and lUD use), and examined tiie men's
abihty to perform and maintam behaviour (self-efficacy in convincing wives to start
or to continue lUD use). The use of measures of decisional balance and self-efficacy
178
for confraception have been reported and vaUdated m otiier sttidies, but mamly for
HIV/AIDS prevention (Grimley et al.l995; Galavotti et al.l995; Grimley and Lee
1997; Stark et al.l998). The review of the literattwe did not reveal any measures for
RJD use at the time the study was designed. Given no existmg measures (scales)
and the need for such measures to be specific for the behaviour (lUD acceptance)
and context (a rural Vietiiam community), the pilot study was conducted to develop
rehable and valid measures for the two rural communes. ReUable and valid
measures for the readiness of rural Vietnamese men to accept lUD use were
developed and vaUdated in the pilot study (Ha et al. in press).
Confraceptive behaviour is influenced by the dyadic relationship between couples.
The literature suggests that husband's approval is the most important factor in
contraceptive use (Joesoef et al. 1988; Lasee and Becker 1997; Kamal 2000) and
husband's approval will increase the odds of using an lUD by 5.42 times (Kamal
2000). In developing the scales for pros and cons, focus group discussion and indepth interviews were conducted to ascertain relevant saUent beUefs. The developed
scale took into account the attitudes and behaviour of a wife who may have been
influencing her husband's lUD acceptance. For example, the cons for lUD use,
which included wives' concem about possible side effects (bleeding, abdominal
pain, etc.), were found as significant predictors for delayed progress from the
contemplation/preparation to the action/maintenance stage. Those who reported less
concem about side effects were more likely to accept the lUD for contraception
than those who reported high concem for the lUD. Thisfindingwas consistent with
other studies in family planning, where contraceptive users tend to report less
barriers than non-users (Lowe and Radius 1987; Keith et al. 1991). Studies also
show that side effects are the main barrier for confraceptive adoption (Bongaart and
Bmce 1995).
TTM, the model for which both 'sfrong and weak principles of progress' govem the
pros and cons constmcts, posits that progress from the precontemplation stage to the
action stage involves approximately 1 standard deviation (SD) increase in the pros
(sfrong principle) and approximately a 0.5 SD decrease in the cons (weak principle)
(Prochaska et al. 1992). Results from tiie baseline survey, along witii the means and
standard deviations of these constmcts across the stages, did not provide direct
179
support for this principle (Table 5.17). However, consistent witiitiiemodel, pros for
lUDs were significantiy (P<0.05) higher for those men m tiie action/mamtenance
tiian those m the precontemplation stage, while cons for lUD use were significantly
(P<0.05) lower for those in the action/maintenance stage than for those in the
precontemplation.
But this direction was not found for pros and cons for
confraception in general
Interpersonal and social/situational aspects that may affect willingness or ability to
accept the lUD for confraception were captured by using tiie constmct of selfefficacy. Self-efficacy for lUD use assessed men's confidence to convince wives to
use, or to continue using, the lUD despite concems about headache, bleedmg,
abdominal pain and difficulty in sexual intercourse. People who were more
confident in convincing wives to use, or to continue using, an lUD were almost 3
tunes more likely to accept an lUD for confraception than those who were less
confident. Thefindingis consistent with other studies. It has been argued that selfefficacy is the most important prerequisite for behavioural change (Bandura 1986),
and it has been studied with respect to prevention of unwanted pregnancies and
shovm to be a significant predictor of confraceptive behaviour (Levinson et al.
1998; Cecil and Pinkerton 1998).
In the baseline study, self-efficacy for lUD use increased from the precontemplation
stage to the action/maintenance stage (Table 5.17), consistent with other TTM
studies. An increase in self-efficacy has been consistently observed in crosssectional studies that show good discrimination between stages, with individuals in
higher stages exhibiting higher self-efficacy than those in lower stage (Velicer et al.
2000).
In both pilot and baseline studies, none of the measures for confraception in general
were found to be significant predictors of men's readiness to accept lUDs for
confraception. This resuh is inconsistent with other studies where pros and cons and
self-efficacy were significantiy associated with stage of change for confraception
(Grimley et al. 1995; Galavotti et al. 1995; Stark et al. 1998).
180
Each confraceptive method requires specific skills and knowledge for its use.
Knowledge and skills for lUD use are very different from those requfred for the use
of condoms or the pill. Women who accept lUDs for confraception need to get the
lUD inserted only once, and they need not worry about giving it any further
technical attention. The use of condoms and the pill, however, requfre continual
preparation and effort such as purchasing the item, getting the condom ready before
intercourse and taking the pill everyday. Moreover, people who miss a dose or two
of tiie pill need to have the knowledge to deal witii this sittiation. Therefore, tiie
intermediate constmcts of TTM (decisional balance and self-efficacy), as measures
for general confraceptive use are not significant predictors of men's readiness to
accept the lUD for confraception, as they do not capture the specific skills and
knowledge for its use.
Maibach and Murphy (1995) recommend three factors that should be considered
when measuring self-efficacy: the behaviour, the specific situation and time frame.
People tend to avoid the task that they believe exceeds their capabiUty. People will
not accept a confraceptive method if they feel the use of the method requires effort
that exceeds their capability. Therefore, in order to assess the readiness to accept a
specific confraceptive method, people need to be asked specific items measuring the
pros/cons and confidence related to that method in particular, not items measuring
confraception in general.
Contraceptive knowledge
Knowledge of a confraceptive metiiod can be measured by assessmg what a
participant recalls spontaneously and what s/he recalls witii prompting (Ezeh et al.
1996). In the study sample, participants' knowledge of confraceptive methods was
very high (99.4 percent) and was similar to levels found in a national health
demographic survey in 1997. In Viettiam, the recall of at least one confraceptive
metiiod was found to be 98.7 percent (NCPFP 1999). The level of recall m the
study was higher (99.4 vs. 96.3 percent) than in another study conducted in
mountain villages (Mai and Montague 1998). The geographic difference and
accessibiUty were given as the main reasons for the differences in knowledge about
confraceptive methods between tiie two geographic regions (NCPFP 1999).
181
Participants in this sttidy had higher knowledge than those in West Afiica (85
percent) and East Afiican (98.8 percent), but were similar to those m a Bangladeshi
sttidy (99.7 percent) (Ezeh et al. 1996). The most widely known metiiod was tiie
lUD, followed by condoms and the pill, while in other countties the best known was
tiie pill, followed by condoms and steriUsation (Ezeh et al. 1996). This reflects tiie
difference in confraceptive methods used in these countiies. In Vietnam, the lUD is
the most commonly used confraceptive metiiod, while m other countries the pill and
condoms are the two most widely used modem methods.
Different knowledge mdicators were measured m the study i.e., spontaneous,
prompted and total recall of modem, fraditional and at least one confraceptive
metiiod. However, only spontaneous recall of fraditional methods was found a
significant predictor of stage of change for lUD use. Thefindingis inconsistent witii
other studies, where knowledge of modem confraceptive methods was reported as a
significant predictor in modem contraceptive use (National Research Council 1993).
Difference by stage may not have found, as there is a high level of confraceptive
knowledge at baseline.
Communication on family planning issues
Couple communication is a significant predictor of confraceptive use. It allows
shared decision-making and more equitable gender roles (Drennan 1998). More
than two-thfrds of couples communicated on family planning in the year that
preceded the baseline survey. More frequent communication was found on topics
such as the use of confraception, sex and family size, and less communication on
spacing of births and abortion. It seems that as more than two-thfrds of men did not
want to have any more children, the question of spacing and abortion did not
deserve attention. The frequency of communication is similar to other studies in
Vietiiam (Mai and Montague 1998; Mai 2001) and hidia (JoUy 1976), but ft was
higher than that in Pakistan (Mahmood and Ringheim 1997).
The first step in a rational process of fertility decision-making in a family involves
communication between spouses. Such communication should be the most
182
important precursor of lower desired family size and mcreased confraceptive use.
Research shows that discussion between partners is a positive predictor of current
confraceptive use (Odimegwe 1999). However, in this sttidy, couple communication
was negatively associated with men's acceptance of the lUD for confraception.
Those people, who had accepted the lUD for contraception for six months or longer
were 1.8 times more likely to have a low level of communication witii wives on
family planning tiian those who had not accepted the lUD. The highest proportion
with a high level of communication (34.8 percent) was found among men belonging
to the contemplation/preparation stage. A lower proportion was reported among
those in the precontemplation stage (27.1 percent).
One possible explanation of this finding is that, once lUD use is established, the
need for discussing family planning is minimal, as it is a reasonably stable method.
Therefore, more communication between the partners on family planning is to be
expected among those in the precontemplation and the contemplation/preparation
groups.
The baseline survey revealed that about 90.0 percent of men did not communicate
with people other than their wives on family planning issues. Despite a long history
of the family planning program, men still were reluctant to discuss such issues with
other people since it is perceived as belonging to 'women's matters'.
In summary, TTM provided a useful framework for understanding how men
intentionally changed their acceptance of the lUD. The model defines change as a
gradual, continuous and dynamic process (Prochaska et al. 1994). The SOC
provides information on when a particular shift in pros, cons and self-efficacy, and
behaviour may occur. Findings from the baseline study confirmed that the pros and
cons and self-efficacy for lUD use appear to be useful measures of readiness to
accept lUDs, with the capacity to discriminate among people in different stages.
People in the action/maintenance stage were more likely to have higher self-efficacy
and lower cons for lUD use than those in the two lower stages. Measures specific
for lUD use were more sensitive to men's readiness to accept the lUD than
183
intervention was deUvered to all people in the same stage, was used (Rakowski et al.
1998). The intervention material and information was provided more than once, on
the premise that behaviour change is best fostered in gradual steps. If participants
changed thefr stage of readiness to accept the lUD for confraception between two
intervention
rounds
(e.g.,
moving
from
the
precontemplation
to
the
correspondmg
to
their
new
stage
of
readmess
to
accept
lUDs
(contemplation/preparation stage).
Some mtervention programs to improve husbands' family planning mvolvement
have also achieved significant results by increasing confraceptive use (Terefe and
Larson 1993; Ozgue et al. 2000); by lowering confraceptive dropout rates (Amatya
et al. 1994); and by reducing abortion rates (Wang et al. 1998). Acknowledgmg tiie
important role of husbands in confraceptive use, this study hypothesized that men
who have received the intervention would be more likely to report higher benefits
(pros) and less costs (cons) for contraception in general and lUD use; be more
confident (self-efficacy) in convincing their wives to use, or continue to use, lUDs;
be more knowledgeable about confraception; be more communicative on family
planning; and report higher lUD acceptance.
Knowledge
In a majority of intervention studies, after exposure to an intervention, a significant
increase in knowledge of reproductive health issues has been reported (Eisen et al.
1992; Speizer et al. 2001; Brieger et al. 2001). In this study, a significant increase
was only seen in recall of fraditional methods for the intervention group. The
intervention letters included messages that identifiedfraditionalmethods and stated
that fraditional methods had low effectiveness for prevention of pregnancy. Those
who were exposed to the letters (intervention group) were therefore more Ukely to
recallfraditionalmethods at the follow-up survey. However, no increase m recall of
modem confraceptive methods was found. This could be due to participants' very
high level of knowledge of modem confraception at the baseline survey (94.0
percent).
Communication
Couple communication has been reported as the most important factor in predicting
confraceptive use (Drennan 1998). It allows people to share beliefs and preferences,
to release fears related to using a method and to find ways to overcome the barriers
that they face in adopting a method. Couple communication increased by 84.0
percent following the communication campaign promoting vasectomy in three Latin
185
American countiies (Vemo 1996). hi another sttidy m Tanzania, women who were
exposed to a communication mtervention were 1.9 times more Ukely to discuss
family plannmg with thefr spouses, and were 1.7 tunes more likely to adopt a
confraceptive method (Jato et al. 1999).
In this study, significant increase in couple communication on family plannmg and
lUD acceptance were reported. Encouragement for men's communication witii
their wives and other people was conveyed in the letters as weU as remforced by the
local health workers. Though the sttidy did not evaluate tiie content of
communication between the husbands and wives, it was assumed that the increase
in the level of communication occurred on topics related to the intervention. The
intervention was designed to increase communication on family planning and lUD
use. Findings showed that the intervention was successful in terms of facilitating
communication between men and their wives and other people on such a private
topic.
TTM constructs: decisional balance and self-efficacy
The pros and cons of the decisional balance constmct are similar to the perceived
benefit and perceived barrier constmcts in the Health BeUef Model (Rosenstock
1974). HBM suggests that individuals weigh potential benefits of the behaviour
against its psychological, physical and financial barriers. In TTM, the perceived
benefits/costs are conceptualised as the pros and cons of the behaviour. The balance
of pros and cons depends on the stage that individuals belong to (Prochaska et al.
1994). In the early stages of this study, men judged that the pros of accepting lUDs
were outweighed by the cons. In the later stages, the opposite pattem occurred. The
crossover took place before the action stage.
Past studies have reported effects for either pros or cons but not both in the same
study. In a quasi-experimental study to assess the impact of adolescent sexual health
interventions in Cameroon, only the effect on lowering perceived barriers to
condom use for the intervention group was reported (Agha 2002). On the other
hand, only the effect for increasing pros (advantages) for condoms was found in an
intervention study among people at HIV risk (Fogarty et al. 2001). The perceived
186
sttidy alerts others to tiie possibility of conductmg mtervention programs witii tiie
active participation of local health workers. These people are experienced m
conductmg health surveys for national health programs as part of thefr routme work,
hivolvement of local health workers helps the researcher wm the confidence of tiie
participants, and obtain tmthful answers to sensitive questions such as those related
to sexual matters. In addition, local health workers were able to verify the accuracy
of tiie study's mformation regarding current lUD acceptance by comparing it to
their own statistics.
hi addition to the above-mentioned strengtiis of the design and hnplementation of
the mtervention, other methodological issues need discussion. Cook and Campbell
(1979:37) defined intemal validity as "the approximate validity with which we can
mfer that a relationship is causal". In a quasi-experimental study, some factors could
be a threat to intemal validity such as equivalence of participants' characteristics,
and extraneous experience or environmental effects.
The equivalence of participants' characteristics refers to the difference between
people who were in the intervention and the confrol groups, as well as the difference
between people who were in the posttest and those who were lost to follow-up. The
best design to obtain two study groups that are equivalent is to randomly assign
people to these study groups; however, it cannot be done in a quasi-experimental
study (Gliner and Morgan 2000). Therefore, the first key question is whether the
intervention and confrol groups were equivalent in all respects prior to the
intervention. To address this question, the study used two methods. Ffrstly, the
baseUne differences were identified and, secondly, the identified baseline
differences were statistically confroUed (using ANCOVA) to adjust for covariate
variables.
In a longitudinal study, as a group of people is followed over time, there is ahnost
always some dropouts or loss of cases. However, in this study, the proportion that
was lost to follow-up was small (less than 7 percent). Therefore, a loss to follow-up
(or mortahty bias) was unlikely to be an issue in the study.
189
A selection bias may have occurred in the sttidy due to non-random selection of
participants. However, any bias was mmunal as the sttidy included all participants
who fulfiUed inclusion criteria from aU the viUages chosen for tiiis study. The
selection of intervention and confrol groups was based on geographic location,
aimed to reduce contMnination between study groups. Acknowledgmg that
contamination cannot be eliminated, the study attempted to reduce its mfluence by
locating the intervention group in villages that were separated by a distance of 2-3
km from the confrol group villages. Nevertheless, some exchange of mformation
between intervention and control villages may have taken place. Communication
between study groups may have affected the performance of the confrol group and,
consequentiy, the resuhs of the study, ft is possible that if participants learn that
they are in the confrol group, they may give-up or not try as hard, exaggerating
differences between the intervention group and confrol groups (Gliner and Morgan
2000). This may be one explanation to the decline of communication in confrol
group after intervention period. The other could be a response bias where the
interviewers took more effort to obtain valid responses from the intervention group
in relation to the confrol group.
As participants become more knowledgeable about the study measures and issues, a
maturation effect can be produced that is independent of the changes, which the
intervention is designed to produce (Gliner and Morgan 2000). People in the
intervention group were intensively exposed to the intervention messages; therefore,
the levels of confraceptive knowledge at posttest were more likely to be better than
at baseline.
Field studies carry with them inherent problems. Many forces can affect the results
of the study. For example, any extemal influence other than the intervention, such
as messages on family planning from radio, television, and new family planning
projects, may have affected results. A monitoring form was designed where all
activities related to family planning in the local areas were recorded. This helped the
researcher to ensure the changes in measures were attributable to the intervention
and not to chance. No additional family planning program or activities other than
routine activities took place in the study areas. There was no major change in the
local health staff in either confrol or intervention villages during the intervention
190
period. Therefore, exfraneous factors were unlikely to have affected the sttidy
results.
Many questions were repeated to the same group of people, and the same local
health workers conducted the baseline and tiie posttest surveys. The mterviewers
were more skilled m the posttest survey than in the baselme survey (as they used the
same questions), and respondents were more likely to be famiUar with the questions
and have prepared answers. However, it is likely that these effects were minimal in
tills study as there was a long interval between baseline and posttest surveys (Gliner
and Morgan 2000).
8.2.5. Limitations
A weakness of the study was that all measures were based on self-report. However,
the reliance on self-report data was necessary because confraceptive behaviour is a
private issue. The self-reported change in social cognitive factors (pros, cons and
self-efficacy) corresponded with the change in observed lUD acceptance. An effect
of over-reporting due to the tendency to give socially desfrable responses was
unlikely to have occurred in the study because the local health workers were
knowledgeable about the confraceptive behaviour of households. The increase in
lUD acceptance in the intervention group at posttest was not significantly due to
biased self-reporting but to actual intervention effects.
A second possible weakness was that all local health workers who conducted the
intervention also evaluated the intervention at posttest. They may have tended to
overestimate the participants' responses. Employment of independent evaluators
blind to the intervention was not possible due to limited resources. However,
supervisors verified the participants' responses and ensured that no systematic overreporting occurred (Chapter 6; Appendix G).
Another possible limitation was that the study did not assess change by each of the
five stages offransition.As presented earlier (Chapter 4), it was decided to compare
the fransitions from the precontemplation to the contemplation/preparation stage
and from the contemplation/preparation to the action/maintenance stage.
191
one intervention component had lower (or no) effectiveness compared witii tiie
overall effectiveness. However, the mam purpose of the study was to test an
mtervention suitable for a mral community in Vietnam. Therefore, tiie issue about
the most effective component of the intervention was not considered to be
important.
8.3. Recommendation and conclusions
This part presents recommendations for further research and improvement to family
planning program in Vietnam.
8.3.1. Recommendations for further research
This study provides additional evidence for the use of behavioural theory-based
interventions in programs promoting the acceptability of a modem confraceptive
method. The findings are consistent with previous studies that showed husband
involvement was likely to resuh in higher effective modem confraception (Terefe
and Larson 1993; Wang et al. 1998), and effectiveness of stage-targeted
intervention in changing behaviour (Velicer et al. 2000).
This research provides the first reported evidence of the efficacy of stage-targeted
interventions for increasing men's acceptance of the lUD for confraception. The
constmcts, self-efficacy and decisional balance, were found useftil in shaping the
contents of targeted letters in the delivery of the intervention.
Further studies could test TTM stage-targeted interventions for other confraceptive
methods such as condoms and the pill in similar rural settings. Condoms and the pill
have been promoted in Vietnam for a long time, but their prevalence is rather low.
Further research to identify the distribution of people by their readiness to accept
these methods is needed. In this study, the five-stage model of change had
questionable appUcation for lUD use. Further research is needed to vaUdate staging
algorithms and time periods (6 months or one year or longer) for lUDs as well as for
other confraceptive methods.
193
Finally, ft is important to view the results of the study in the context of otiier current
social-cognitive models of health behaviour change. Research evaluating the ability
of other theories and models to explain health behaviour change (e.g. confraceptive
behaviour change) is also needed. Research that integrates consttiicts from other
theories and models is worthy of investigation.
8.3.2. Impact of the study on family planning services
Contraceptive use in Vietnam is dominated by the lUD, which accounts for twothfrds of modem confraception. On the other hand, there is a high rate (25 percent)
of use of fraditional methods despfte their high failure rate (NCPFP 1999). The use
of other methods like the pill, condoms and female sterilization are low, at 5- 6
percent. The use of injectables, male sterilization and vaginal methods is negligible.
Vietnam is among the countries with the highest abortion rate; 2.5 abortions per
woman life time (Goodkind 1994) and a substantial proportion of women uses
abortion as a substitute for contraception (Hieu et al. 1993). As promotion of lUD
use targets women, there is lack of attention given to men and to their role in family
planning acceptance. Therefore, it is suggested that a sfronger focus on men would
be an effective sfrategy. The present population policy on male participation in
reproductive health and family planning programs should be sfrengthened.
Couple communication is reported to be the most important factor in confraceptive
use (Drennan 1998). In rural settings, communication occurs but, because men are
seen as the main decision-makers (Nustas 1999; Becker and Costenbader 2001) and
the husbands' confraceptive preferences are likely to prevail (Mason and Smith
2000), couples may not reach a decision together. This study provides empirical
evidence of the low level of communication between the men and thefr spouses.
Using stage-targeted messages informing recipients of the benefits of the lUD and
importance of couple communication feasibly shifted the communication from a
194
low to a high level. The resulting increase in confraceptive use in the mtervention
group suggests that providing men witii more information on tiie unportance of
communication on family planning, helpmg them overcome culttu-al barriers on son
preference, providing counsellmg on the risk of unwanted pregnancy and abortion,
would help them be more supportive and more aware of tiie concept of shared
decision-making in family planning (Wells 1997).
In the case of lUD use, husbands may be unaware of the metiiod, unconcemed
about the side effects and therefore not ready to accept the lUD for confraception
(precontemplation group). A screening insttiiment based on a stage of change
algorithm would be a practical tool that family planning counsellors could use in
field situations in Vietnam. Ashort form of the instrument with few practical and
easy-to-use questions would be helpful for family planning counsellors. The
fraining course for those counsellors on how to identify the men in different stage of
readiness to accept lUD, with role-play as conducted in this study is a practical
method to implement this change.
Once identified, the precontemplation group should be provided with basic
information about lUDs, i.e. how to use it, its potential side effects and how to
manage side effects. When men become more conscious of the above issues and
decide to use the lUD, they should be provided with messages that help to increase
their confidence (self-efficacy) in persuading their wives to accept lUD use
(contemplation/preparation group). The message given to the men should be short
and easy-to-understand, using local language. The study has provided additional
empirical evidence of the value of increasing men's self-efficacy in increasing
wives' acceptance of the lUD (Kim et al.l998).
From a service delivery perspective, findings from the study have important
implications for the efficient and effective delivery of family planning programs in
rural areas. The challenge for family planning in Vietnam is not only to promote
contraceptive-mix by increasing the use of modem methods such as condoms and
the pill. The adoption of such an approach would be helpful in implementing
family planning programs in Vietnam but it is also necessary to target men to
195
196
REFERENCES
Amatya, R., Akhter, H., WilUamson, N., Ahmed, Y., McMahan, J. and Gates, D.
(1994), The effect of husband counselling on NORPLANT
confraceptive
An Hai Disttict Health Cenfre (2001), Annual Report. An Hai Disttict Health
Cenfre. An Hai, Hai Phong. Vietnam.
Angle, M. A., Berfrand, J. T., Hardee, K. and Magnani, R. J. (1995), Access, quality
of care and medical barriers in family planning programs. Intemational Family
Planning Perspectives 21, 64-69.
Anh, H. T., Minh, T. H., Due, N. M., Mai, B. T., Thien, P. V., Ha, V. S., Ha, P. T.,
Trang, Q. T., Schuler, S. R. and SantiUan, D. (2002), Gender, Empowerment of
Women Status. Population Development Intemational, Hanoi. Vietnam.
Amold, F. (1992), Sex preference and its demographic and health implications.
Intemational Family Planning Perspectives 18(3), 93-101.
197
Bankole, A. and Singh, S. (1998), Couples' fertility and confraceptive decisionmaking in developing countiies: hearing the man's voice. Intemational Family
Planning Perspectives 24(1), 15-24.
Belenger, D. and Hong, K. T. (1998), Young single women using abortion m Hanoi,
Vietnam. Asia -Pacific Population Journal 13(2), 3-26.
Bongaarts, J. (1991), The KAP- Gap and tiie unmet need for confraception.
Population and Development Review 17 (2), 293-313.
198
Bongaarts, J. and Bmce, J. (1995), The cause of unmet need for confraception: a
social content of services. Studies in Family Planning 26 (2), 57-75.
Brieger, W. R., Delano, G. E., Lane, C. G. and Oladepo, O. (2001), West Afiican
Youth Initiative: outcome of a reproductive health education program. Joumal of
Adolescent Health 29,436-446.
Bryan, A. D., Aiken, L. S. and West, S. G. (1997), Young women's condom use: the
influence of acceptance of sexuality, confrol over the sexual encoimter, and
perceived susceptibility to common STDs. Health Psychology 16(5), 468-479.
Bryant, R., Ross.J., Bhushan, I., Blackburn, R. and Sherman, J. (1996), Meeting
unmet need: new sfrategies. Population Report. J (43).
Cabral, R. J., Galavotti, C , GargiuUo, P. M., Amfrong, K., Cohen, A., Gielen, C. A.
and Watkinson, L. (1996), Paraprofessional delivery of a theory based HIV
prevention counselling intervention for women (Special issue: behavioural science
in HIV prevention). Public Health Reports 111, 75-82.
199
47,113-123.
Cohen, M. (2000), Trauma ward. Far Eastern Economic Review. 29, June.
200
Doll, J. and Orth, B. (1993), The Fishbein and Ajzen theory of reasoned action
appUed to confraceptive behaviour. Model variants and meanfrigfiihiess. Joumal of
Applied Social Psychology 23(5), 395-415.
Donatti, S., Hamam, R. and Medda, E. (2000), Family plannmg KAP survey m
Gaza. Social Science and Medicine 50, 841-849.
Edlund, L. (1999), Son preference, sex ratios, and marriage pattems. Joumal of
Political Economy 107(6), 1275-1295.
Eisen, M., Zelhnan, G. L. and McAlister, A. L. (1992), A healtii behef modelsocial learning theory approach to adolescents' fertiUty confrol: findings from a
confroUed field tiial. Health Education Quarterly 19(2), 249-262.
201
Entwisle, B., Rindfiiss, R. R., Walsh, J. S., Evans, T. P. and Cuiran, S. R. (1997),
Geographic mformation system, spatial network analysis, and confraceptive choice.
Demography 34{2), 171-187.
Fishbein, M., Higgins, D. L., Rietmeijer, C. and WoUtski, R. J. (1999), Communitylevel HIV intervention in 5 cities: final outcome data from the CDC AIDS
community demonsfration projects. American Journal of Public Health 89(3), 336345.
Fisher, A. A., Laing, J. and Stoeckel, J. (1985), Guidelines for overcoming design
problems in family planning operations research. Studies in Family Planning 16(2),
100-105.
Fogarty, L. A., Heilig, CM., H., Amsfrong , K., Cabral, R. J., Galavotti, C , Gielen,
C. A. and Green, B. M. (2001), Long term effectiveness of a peer-based
intervention to promote condom and confraceptive use among HIV-positive and dirisk women. Public Health Reports 116(1), 103-123.
202
Gage, A. J. (1998), Sexual activity and confraceptive use: the components of the
decision making process. Studies in Family Planning 29 (2), 154-167.
Gallen, M. E. and Liskin, K. N. (1986), Men- new focus for family planning
programs. Population Reports Series N 33.
Gerrard, M., Breda, C. and Gibbons, F. (1990), Gender effects in couples sexual
decision making and confraceptive use. Journal of Applied Social Psychology 20
(6), 449-464.
Glanz, K., Rimer, B. K. and Lewis, F. M. (1997), Health Behaviour and Health
Education: Theory, Research, and Practice. Second Edition. Jossey-Bass, San
Francisco.
203
Goodkind, D. (1995.), Vietnam's one or two child policy in action. Population and
Development Review 21 (1), 85-112.
Gorbach, P. M., Hoa, D. T. T., Nhan, V. Q. and Tsui, A. (1998), Confraception and
abortion in two Vietnamese communes. American Joumal of Public Health 88,
660-663.
Graham, M. J., Larsen, U. and Xu, X. (1998), Son preference in Anhui province,
China. Intemational Family Planning Perspectives 24(2), 72-77.
Grimley, D. M. and Lee, A. P. (1997), Condom and other confraceptive use among
a random sample of female adolescents: a shapshot in time. Adolescence 32 (128),
771-779.
204
Haughton, J. (2000), Ten Puzzles and Surprises Economic and Social Change m
Vietiiam, 1993-1998. Comparative Economic Studies 42{4), 67-80.
Hester, N. R. and Macrina, D. M. (1985), The healtii beUef model and tiie
confraceptive behaviour of college women: implications for health education.
College Health 33, 245-252.
Hicks, D. A. (1998), What risk of infection witii lUD use? The Lancet 351(9111),
1222-1224.
Hieu, D. T., Van, H. T., Donalson, P. J. and Nga., Q. L. (1995), The pattem of lUD
use in Vietnam. Intemational Family Planning Perspectives 121, 6-10.
205
Hussain, R., Fikree, F. F. and Berendes, H. W. (2000), The role of son preference m
reproductive behaviour m Pakistan. Bulletin of the World Health Organization
78(3), 379-386.
Jahan, S. A., Thwin, A., Tunon, C, and Nasreen, S. (1996), Urban Men and their
Participation in Modern Contraception. An Exploratory Study. MCH-FP extenstion
project (urban). Health and population extension division. ICDDR. Bangladesh.
206
Johansson, A., Hoa, H. T., Tuyet, L. T. N., Bich, M. H. and Hojer, B. (1996),
Family planning in Vietnam - women's experiences and dilemmas: a community
sttidy from the Red River Deha. J Psychologmo. Obstet. Gynecol 17, 59-67.
Johansson, A., Lap, N. T., Hoa, H. T., Diwan, D. W. and Eriksson, B. (1998a),
Population pohcy, son preference and the use of lUDs in North Vietnam.
Reproductive Health Matters 6(11), 66-76.
Johansson, A., Nga, T. N., Huy, T. Q., Dat, D. D. and Hohngren, K. (1998b),
Husbands' involvement in abortion in Viettiam. Studies in Family Planning 29(4),
1-13.
Karra, M. V., Stark, N. N. and Wolf, J. (1997), Male involvement in family planing:
a case study spanning five generations of a South Indian family. Family Planning
Perspectives 28(10), 24-34.
207
Keith, J. B., McCreary, C , Collins, K., Smith, C. P. and Bemstem, I. (1991), Sexual
activity and confraceptive use among low-income urban blacks. Adolescence
26(104), 769-785.
Kim, Y. M., Kols, A. and Mu, S. (1998), Informed choice and decision-making in
family planning counselling in Kenya. Intemational Family Planning Perspectives
24, 4-6.
Kmg, A. C , SalUs, J. F., Dunn, A. L., Simons, M. R. D., Albright, C.A., A., Cohen,
S., Rejeski, W. J., Marcus, B. H. and Coday, M. C. (1998), Overview of the activity
counselling trial (ACT) intervention for promoting physical activity in primary
health care settings. Medicine and Science in Sports and Exercise 30 (7), 1086-96.
Knodel, J., Anh, P. T. and Vinh, D. X. (1995), Vietnam's Population and Family
Planning as Viewed by its Implementers. Population Council, Hanoi.
Kreuter, M., Farrell, D. W., Olevitch, L. and Brennan, I. (1999), Tailoring Health
Messages: Customizing Communication With Computer Technology. Lawrence
Erlbaum Assoc.
208
Family Planning
Levin, A., CaldweU, B. and Khuda, B. (1999), Effects of price and access on
contraceptive use. Social Science and Medicine 49, 1-15.
Levinson, R. A., Wan, C. K. and Beamer, L. J. (1998), The confraceptive self efficacy analysis in four samples. Journal of Youth and Adolescence 27 (6), 773793.
Lucker, K. (1975), Taking Chances: Abortion and the Decision not to Contracept.
University of Califomia Press, Berkeley, CA.
Mai, T. T. P., Nam, T.H., and An, D.T. (2001), Male Role in Family Planning in
Thai binh and Hoa binh Provinces. Ministry of Health. MCH/FP Department.
Hanoi.
Malotte, C. K., Jarvis, B., Fishbein, M. and Kamb, M. (2000), Stage of change
versus an integrated psychological theory as a basis for developing behaviour
change intervention. AIDS Care 12(3), 357-364.
Mason, K. O. and Smith, L.H. (2000), Husband's versus wives' fertihty goals and
use of confraception: the influence of gender context in five Asian countties.
Demography 37(3), 299-311.
May, T. (1993), Social Research: Issues, Methods, and Process. Open University
Press, Buckingham. Philadelphia.
Mbizvo, M. T., Kasule, J., Gupta, V., Rusakaniko, S., Kinoti, S. N., Shumbushu, W.
M., Zziwa, A. J. S., Mwateba, R. and Padayachy, J. (1997), Effects of a randomized
health education intervention on aspects of reproductive health knowledge and
210
Mensch, B., Kuenning, M. A. and Jain, A. (1996), The impact of quaUty of family
planning services on confraceptive use in Pern. Studies in Family Planning 27(2),
59-75.
Miller, W. B. and Pasta, D. (1996), The relative influence of husbands and wives on
the choice and use of oral confraception, a diaphragm, and condoms. Joumal of
Applied Social Psychology 26 (19), 1749-1774.
Mueller, R. D. and Germain, A. (1992), StaUdng the elusive "unmet need" for
family planning. Studies in Family Planning 23 (5), 330-335.
MOH (2002), The policy and solutions in implementing primary health care
strategies. Ministry of Health, Hanoi.
211
National Research Council (1993), Factors affecting contraceptive use in subSaharan Africa. National Academy Press, Washmgton, D.C.
NCPFP (1999), Vietnam Demographic and Health Survey 1997. In Depth Analysis
for Population and Family Health Baseline Values. National Committee for
Population and Family Planning, Hanoi, Viettiam.
NCPFP and GTZ (1995), Reproductive Health Survey 1995. NCPFP and GTZ.
Hanoi.
Necchi, S. (1999), Men, Family Formation and Reproduction. PoUcy and Research
Paper. http://www.iussp.org/Publications_on_site/PRP/prpl7.htm.
Ngom, P. (1997), Men's unmet need for family planning: impUcation for Afiican
fertility fransitions. Studies in Family Planning 28(3), 192-202.
Nhan, V. Q., Thom, N. T. and Ross, A. J. (1999), Contraceptive Method Mix study
in Vietnam. The Results of Second Survey. The Future Group Intemational. NCPFP.
The Popopulation Council. Hanoi.
Nhan, V. Q., Mai, N. T. P., Hau, N. T., Miller, R. A., Kane, T. T., Stoeckel, J.,
Long, L. D., Ha, B. T. T., Thi, C. P. and Thom, N. T. (2000), A Situation Analysis
of Public Sector Reproductive Health Services in Seven Provinces of Vietnam.
Population Council. Hanoi.
212
Noble, J. (1996), Family plannfrig choice still lags m Vietiiam. Population Today
24 (1), 96-97.
Nustas, W. P. (1999), Men's knowledge of and attitudes toward birth spacmg and
confraceptive use in Jordan. Intemational Family Planning Perspectives 25(4), 181185.
Ozgue, S., Disan, B. and Ozcirpici, B. (2000), The effects of family planning
education provided to different gender groups. British Joumal of Obstetrics and
Gynaecology 107(10).
Pallant, J. (2001), SPSS Survival Manual. A Step by Step Guide to Data Analysis
Using SPSSfor Windows (Version 10). Allen and Unwin, AusfraUa.
213
Phai, N. v., Knodel, J., Cam, M. V. and Xuyen, H. (1996), Fertihty and family
planning m Vietnam: Evidence from the 1994 Inter-censual Demographic Survey.
Studies in Family Planning 27(1), 1-17.
Prochaska, J. O., VeUcer, W. F., Rossi, J. S., Goldstem, M. G., Marcus, B. H.,
Rokowski, W., Fiore, C , Harlow, L. L., Redding, A. C , Rosenbloom, D. and Rossi,
S. (1994), Stages of change and decisional balance for 12 problems behaviours.
Health Psychology 13 (1), 39-46.
214
Rakowski, W., Ehrich, B., Goldstem, M. G., Runer, B. K., Pearhnan, D. N , Clark,
M. A., Vehcer, W. F. and Woolverton, H. (1998), hicreasmg mammography among
women aged 40-74 by use of a stage-matched, tailored intervention. Preventive
Medicine 27, 748-756.
Ross, J. and San, P. B. (1997), Unmet need in Vietnam: who needs what and when.
Social Biology 44 (1-2), 111-13.
Salway, S. (1994), How attitudes toward family planning and discussion between
wives and husbands affect confraceptive use in Ghana. International Family
Planning Perspectives 20(2), 44-47.
215
Speizer, I. S. (1999), Are husbands a barrier to women's family planning use? The
case of Morocco. Social Biology 46{\-2), 1-16.
Stark, M. J., Tesselaar, H. M., O'Connel, A. A., Person, B., Galovotti, C , Cohen, A.
and Walls, C. (1998), Psychological factors associated witii the stages of change for
condom use among women at risk for HIV and STDs: implications for intervention
development. Joumal of Consulting and Clinical Psychology 66, 967-978.
Stash, S. (1996), Ideal family size and sex composition preference among wives and
husbands in Nepal. Studies in Family Planning 27(2), 107-119.
216
Thom, N. T., D. T. Hieu, Nhan, V.Q., Thanh, D.N., Whittaker, M., Fajans, P., Satia,
J. (2000). "The sfrategic approach to the infroduction of DMPA as an opportunity to
quality of care for all confraceptive methods in Vietnam." Asia -Pacific Population
Joumal 15(4): 64-88.
Toan, N. V., Hoa, H. T., Trong, P. V., Hojer, B., Persson, L. A. and Sundsfrom, K.
(1996), Utilization of reproductive health services in rural Vietnam: are there equal
opportunities to plan and protect pregnancies?. Joumal of Epidemiology and
Community Health 50(4), 451-455.
217
Tmssell, J., Leveque, J. A., Koenig, J. D. and London, R. (1995), The economic
value of confraception: a comparison of 15 methods. American Joumal of Public
Health 85(4), 494-453.
Tu, P., Qui, S., Fang, H. and Herbert, L. S. (1997), Acceptance, efficacy, and side
effects of Norplant frnplants in four counties in North China. Studies in Family
Planning 28(2), 122-132.
UNDP (1995), Gender kit. Briefing KIT. The Viettiam Women's Union. 2000.
Hanoi, Vietnam
United Nations (1999), Health Sector Review. A United Nations common country
assessment of Vietnam. Hanoi, Vietnam.
VeUcer, W. F., Prochaska, G. E., Fava, L. J., Rossi, J. S., Redding, A. C , Laforge,
G. R. and Robbins, L. M. (2000), Using the Transtheoretical model for populationbased approach to health promotion and disease prevention. Homeostasis in Health
and Disease 40, 174-195.
Verme, C. S., Wegner, M. N. and Jerzowski, T. (1996), What do you mean by that?
Populi
23{2),\0-\2.
218
Wegner, M. N., Landry, E., WiUcinson, D. and Tzanis, J. (1998), Special report.
Men as partners m reproductive health: from issues to action. Intemational Family
Planning Perspectives 24(1), 1-9.
Yount, K. M., Langsten, R. and Hill, K. (2000), The effect of gender preference on
confraceptive use and fertility in rural Egypt. Studies in Family Planning 31(4),
290-300.
219
220
- ^
^^
F
^
e^
Vo^
O
cs
^
-!
^I
OS
221
University of Wollongong
EXPEDITED REVIEW
In reply please quote: SD:KM HEOO/198
Further Enquiries: Karen McRae (PH: 42214457)
7 November 2000
Mrs B. Ha
Graduate School of PubUc Health
University of WoUongong
Dear Mrs Ha,
1 am pleased to advise that the foUowing Human Research Ethics appUcation has been
approved. As a condition of approval, the Human Research Ethics Committee requfres that
researchers immediately report anything which might warrant review of ethical approval of the
protocol, including: serious or unexpected adverse effects on participants, proposed changes to
the protocol, unforseen events that might affect continued ethical acceptabUity of the project
and discontinuation of the research project before the expected date of completion.
Etiiics Number:
HEOO/198
Project Titie:
Name of Researchers:
Mrs B. Ha
Approval Date:
1 November 2000
Duration of Clearance:
31 October 2001
Please note that experiments of long duration must be reviewed annually by the Committee and
it will be necessary for you to apply for renewal of this application if experimentation is to
continue beyond one
University of Wollongong
rNEWAL
(\ reply please quote: SD:KM HEOO/198
further Enquiries: Karen McRae (PH: 42214457)
31 July 2001
Mrs B. Ha
Graduate School of Public Health
University of Wollongong
Dear Mrs Ha,
HEOO/198
Project Title:
Name of Researchers:
Mrs B. Ha
Approval Date:
27 July 2001
Duration of Clearance:
26 July 2002
Please note that experiments of long duration must be reviewed annually by the
Committee and it will be necessary for you to apply for renewal of this
appjjcation if experimentation is to continue beyond one year.
P . s . P^^^a S ^
\l
iCXr^
CA. C O
rt^
Office of Research
.^^riivs-vOK-^^
c ^ aiNfN-Najvr
^,Oe^^or.o.c:^W^^
office research@uow.edu.au
www.uow.edu.au
l^'-atm'iVV-
University of Wollongong
RENEWAL
In reply please quote: SD:KM HEOO/198
Further Enquiries: Karen McRae (PH: 42214457)
9 September 2002
Mrs B. Ha
Graduate School of Public Health
University of Wollongong
Dear Mrs Ha ,
1 am pleased to advise that renewal of the following Human Research Ethics
application has been approved. As a condition of approval, the Human Research
Ethics Committee requires that researchers immediately report anything which
might warrant review of ethical approval of the protocol, including: serious or
unexpected adverse effects on participants, proposed changes to the protocol,
unforseen events that might affect continued ethical acceptability of the project
and discontinuation of the research project before the expected date of
completion.
Ethics Number:
HEOO/198
Project Title:
Name of Researcher/s:
Mrs B. Ha
5 September 2002
Duration of Renewal:
30 December 2002
Office of Research
office_research@uow.edu.au
www.uow.edu.au
17/10 '00
09:25
8 4 4 8452738
SOPH
giooi
OctobeFl3,2000
Chairperson
Human Research Ethic Committee.
Universtify of WoHongong. NSW 2522, Australia.
Tel: (02) 4221 3386
LEVUANH. Ph.D.
Dean ojf Hanoi School of Public Health.
Chairperson, Human Research Ethic commrttee.
222
TO BE READ IN CONJUNCTION WITH THE UNTVERSTTY OF WOLLONGONG HUMAN RESEARCH ETHICS APPUCATION
GUIDELINES
CONSENT FORM
MALE INVOLVEMENT IN FAMILY PLANNING IN RURAL VIETNAM
I
have read the Male involvement in family planning in
rural Vietnam study information sheet. I understand that participation in the study is
volvintary and that I am free to terminate the interview at any stage, I am aware that the
interview will be recorded and that this recording will remain confidential. I am aware that
recorded questionnaire wiU be coded to ensxxre my anonymity.
I understand that the interview will take approximately 30 minutes and will take place at a
time and location at are suitable for me. I am aware that the interview has no bearing on m.y
current or future life.
I understand that fhe infonnation gained from the interview will be used to write Ph.D
thesis and that the information may be published in appropriate journals and in conference
proceedings.
I have discussed with Mrs Bui Thi Thu Ha my participation in the study and I have had the
opportunity to ask questions.
I am aware that I can contact either Mrs Bm Thi Thu Ha (84 4 8452 822) or Dr Le Vu Anh (84
4 8452 822) if I have any further questions relating to the study.
If I have any concems or complains about the conducts of the study I will contact the Ethics
Officers at the Uruversity of Wollongong on 02 4221 3386, Dr Rohan Jayasiiriya on 02 4221
3344 and Hanoi School of Public Health; Dr Le Vu Anh on 84 4 8452 822
I agree to participate in this study.
Signature
Date
Print name:
13
223
TRAINING PROGRAM
This is an intervention study, which is aimed to evaluate the role and contribution
of men in family planning. The more activly men participate in family planning,
the more effective contraceptive use will be reported, particularly in reference to
lUD use. The study used the transtheoretical model of behaviour change as the
underlying framework.
Time: 2/2001-7/2002.
Stages
1- Pilot study: 2/2001.
2- Baseline: 3-4/2001.
3- Intervention: 11/2001 -6/2002
4- Posttest: 6-7/2002.
Location
An hai- Hai Phong
Two communes, Quoc Tuan and An Hong, and 6 villages from each commune
were selected
1.
Commune:
2.
Village:
3.
4.
5.
Household
number:
Name of men
Group
6.
ID number
4 NhuKieu
5 Nhu Kieu
6 Kieu
thuong
An Hong (2)
1 Pham Dung 4 Le Lac 1
5 Ngo
2 Tat xung
Hung
3 Le Lac 2
6 Lesang
Intervention 1: 1.3 (CI); 1.4 (C2); 1.6 (C3); 2.2 (C4); 2.5
(C5); 2.6 (C6)
Confrol 2: 1.1 (Dl); 1.2 (D2); 1.5(D3); 2.1 (D4); 2.3 (D5);
2.4 (D6)
C V G MO^[TH YEAR BIRTH
TTie second column (village), look at the identification part to know the code of each
village. For example, if the participant comes from Van xa village, then number 3 will
be written in the second column next to the first column of number ' 1'
1
Third column (study group). There are two study groups, intervention group and
confrol group. The abbreviation for the intervention group is C and D for the control
group. If the men come from Quoc Tuan commune. Van xa village, then the number
should be 1.3. C. 1.
|3
1
"
11
The next column will record the name of participant. Each colunm records only one
letter. Remember to write only name, not sumame nor middle name. For example, if
the name of participant was Nam, then record only NAM in the next three columns
N
M
The last two columns are for recording the year of birth. For example, if the year was
1957, then the 57 will be recorded in the last two columns.
1
1 N
M
Results of interview
This part is designed for interviewer to use after finishing interview. In addition to
writing the name and date of interview, it is also necessary to summarise the work that
they have done in this interview.
-
If the interview was completed, circle the number 1. If the participant was absent,
circle the number 2. If the participant was not available, and the interviewer needs
to come back to do the interview, cfrcle the number 3. If the man refuses to
answer the questions, circle the number 4. For any other cases, please write in the
result box.
MONTH: [ 1 / I J
DK
98
NO RESPONSE
99
YEARL5 / 7 J
DK
98
NO RESPONSE
99
Q 102 This question is to re-check the accuracy of answers because many people may
give the date according to the lunar calendar which will cause a difference of one
year.
Q102
The person who is cleaning data will be responsible for identifying accurate answer.
Q 103 - Q 105: these three questions ask about the educational level of participants
Q 103: If the man answers that he did not study at all,tiienskip to question Q 106,
and record in tiie respective space the number '2'. If tiie anwer is yes, then record tiie
number ' 1' and ask next question.
Q103
YES 1
NO 2
NO RESPONSE
99
> Q 106
Question Q104 asks about the educational level. Please record the respective number
of educational level. For example, if the educational level was lower secondary
education, then record number ' 3 '
Q104
ILLriERATE 1
PRIMARY
2
LOWER SECONDARY 3
UPPER SECONDARY 4
HIGHER
5
DON'T KNOWN
98
NO RESPONSE
99
Q 105: this question asks the participant to estimate total years of education that he
has had. Here are some possible estimations:
Primary education: 5 years
Lower secondary education: 4 years
Upper secondary education: 3 years
Higher education: need to check how many years in university or college. For
example:
College: 2 years
University: 4- 6 years
Postgraduate education: 2 - 3 years
Then the interviewer needs to write up the total years in the respective box. For
example, ifthe participantfinishedonly lower secondary education, then the total
years of education will be 9, so write the number '9' in the box
Q105
Some people do not remember tiie total years and do not answer tiiis question; when
this happens,cfrcle the number '99'
Q 106: this question askstiieparticipant about the lengtii of time he has lived in the
commune. If he has livedtiierelesstiianone year, tiien circletiienumber '00'. If he
does know or remember, tiien circletiienumber 98; if he does not answer tiie
question, tiien circle tiie number 99. For example, if he has lived about 30 years m tiie
commune, tiien writetiienumber '30'. Do not counttiietimetiiathe lived away from
the commune.
NUMBER OF S L3/0]
IF LESS THAN! YEAR 00
DK
98
NO RESPONSE
99
Q 107: this question asks the participant about his religion. Ifthe answer is different
from religions given here, then drcle 'other'; if 'no rehgion', tiien cfrcle tiie number
'0'. If he does not anwer, then cfrclethe number '99'
Q107
BUDDIST 1
CATHOLIC 2
CHRISTIAN 3
OTHER (SPECIFY)
NO RELIGION 0
NO RESPONSE
99
Q 108:tiiisquestion asks the participant about his ethnitcity. If it is mixed, then circle
the number 8.
Q108
KINH 1
HOA 2
MUONG 3
NUNG 4
HMONG 5
OTHER (SPECIFY)
MIXED ETHNICrTY 8
NO RESPONSE
99
Q 109: this questionasks the participant about his age when he married his current
wife. If he does not remember, circle number 98; if he does not anwer, cfrcle number
99. Please ask the years by Westem calendar, do not ask year by limar calendar. For
example, if he married when he was 24, then write number 24 in the box.
Q109
Q 110: Ifthe man has more than one wife, then ask the age when he married his first
wife. If he has only one wife, then the year should be similar to Q109. Please ask age
by Westem calendar
QUO
GOVERNMENT STAFF
1
PRIVATE BUSINESS
2
AGRICULTURAL /FARMING JOB 3
HOUSE-HUSBAND
4
OTHER (SPECIFY)
Joint-venture company
DK
98
NO RESPONSE
99
Q 112: This question asks the participant about total number of children (living and
dead) tiiat he has ever had. For example, ifthe total number of births was 3, then write
the number *3'
Q112
TOTAL BIRTHS: 3
NONE
0
DK
98
NO RESPONSE
99
Q 113: this question asks the participant about the total number of children that he
hashad with his current wife (sons, daughters, and total children). If there is one
daughter, one son, then the total will be two.
Q113
TOTAL SONS: 1
TOTAL DAUGHTERS 1
TOTAL CHILDREN
2
NONE
0
DK
98
NO RESPONSE
99
Q 114: this question asks the participant about the time of his the last birth,
estimated by year. For example, ifthe last birth was 5 year ago, then record number 5.
If he also knows the month, then record this, for example, 5 years and 6 months. The
month could be rounded up.
Q114
YEARS
MONTHS:
DK
NO RESPONSE
5
6
98
99
Q 115: this question asks the participant about his wife's abortion history. If she has
had an abortion, then circle number 1. If she has had no abortions, then cfrcle number
2. If he does not know, or remember, tiien skip to Q 117. If his wife has had an
abortion, then ask next question Q 116.
Q115
YES
NO
DK
NO RESPONSE
1
2
98
99 >
Q117
Q 117- Q 123: these questions asks the participant forinformation about his wife,
similar to tiie information asked about him. Please refer to instinctions for Q 101-104107-108-109. Remember to record year by Western calendar
Q 202: this question asks the participant about confraceptive metiiods. Ask him
whether he has ever heard about confraceptive methods. If he has, then circle the
number 1.
Q202
YES 1
NO 2
-^ Section
5
Q 203: This question asks the participant whether he knows of any confraceptive
method. If he does not, then ask part 4. Ifthe answer is yes, then cfrcle number 1.
YES 1
Q203 Do you know any family
NO 2
planning method?
Section 3
Q 204: This question asks the participant to list the name (no prompt) of a
confraceptive method. Ifhe does not remember or does not answer, cfrclethe
number 98 or 99 and ask question Q 205
Q204
PILL 1
CONDOMS 2
njD 3
VASECTOMY 4
TUBECTOMY 5
INJECTABLES 6
RHYTHMS 7
WITHDRAWAL 8
OTHERS (SPECIFY)
DK
98
NO RESPONSE
99
For example, ifhe can list two methods, e.g. oral pill and lUD, then cfrcle numbers 1
and 3
Q 205: In the participant does not give the name of any method himself, thenyou
need to promt him for the answer, and circle the method that he remembers in the
respective box. For example, after prompting, he may remember the condom method;
record the number 2 in the respective box.
Q 206 asks the participant about the sources of information. The answers could be
multiple. A similar approach is applied to Q 207.
10
NEVER
0
Yes
1-3 times 1
> 3 times 2
DK
NO RESPONSE
The same approach is used to ask the other questions in this part.
98
99
11
Total number
2
Other
Q 405: this question asks the participant how many sons and how many daughters he
desfres, and whetiier sex makes any difference. For example, ifhe desfres one
daughter and one son, then write the number 1, and 2. Ifhe does not know, circle
number 98; and ifhe does not answer, then cfrcle number 99.
Q405
HAVE A SON 1
HAVE A DAUGHTER 2
NO DIFFERENCE
HAVE BOTH SONS AND
DAUGHTER 3
OTHER (SPECIFY)
DK
NO RESPONSE
98
99
Q 406: this question asks the participant whether he wants to have any more children.
If yes, then write number 1; if no, then write number 2. Ifthe answer is different, then
skip to question Q 408.
Q406
98
99
-> Q 408
12
Q 407: if the participant wants to have another child, then ask how longfromnow
that he wants to have tiie next child. For example, ifhe wants to have a child in the
next 2 years, then write the number 2.
Q407
>2YEARS
1
=<2YEAR
2
CAN'T GEf PREGNANT 3
OTHER SPECIFY
DK
NO RESPONSE
98
99
Q 408: this question asks the participant about his last birth intention, to see whether
he wanted to have the last child or not. Ifthe answer is yes,tiiencfrcle the number '1',
ifhe wanted to have the child later, then cfrcle number 2; ifhe did not want to have
the last child at all, then cfrcle number 3.
Q408
Q409
THEN 1
LATER 2
NOT AT ALL 3
< 6 months
=> 6 months
AC 1
M 2
Q 409: this question asks the participant ifhe is currently using an lUD. Re-check
with Q 402 to see whether the answer is consistent. Ifthe answer is inconsistent, then
ask Q410. Ifthe participant's wife has used lUD for less than 6 months, then cfrcle
number 1; if they have used it for more than 6 months, then circle number 2.
Q410
YES C I
NO
PC 2
->Q411
-> Section 5
Q 410: ifthe participant is currently not use lUD, then ask him ifhe and his wife
intend using an lUD in the next 6 months. If they are intending to use lUD, then cfrcle
number 1 and ask question Q411. Ifthe answer is no, then cfrcle number 2 and move
to the part 6.
Q411
YES PRl
NO C 2
Q 411: this question asks tiie participant who is intending to use lUD intiienext 6
montiis whetiier he is planning to use it intiienext 30 days. If tiiis is his plan, tiien
cfrcle number 1; if not then circle number 2.
13
Q501
Q502
Q503
Q504
important
2
Very
important
know
Benefits of contraception
How important is each of these advantages to you in deciding whether or not to use family
planning methods for prevention of pregnancy using 5 point scale
FP helps you to be more
responsible in decision about
having children
FP helps you avoid the results
of unwanted pregnancy
Your wife would not have to
worry about becoming pregnant
if using contraception
FP helps to limit size of the
family
If the man can choose himself the importance, the mterviewer will record in
the respective box. Ifthe man cannot choose himself the number, then the
interviewer needs to ask the man again, ask him to choose himself the level
between 1 and 5. This is the basic principle for aU questions for part 5 and 6.
After reading tiie question, tiie interviewer needs to summarise tiie ideas behind
tiie question. The main idea of tiie statement is printed Bold and Italic, to make it
easier for the interviewers to summarise the ideas of statement.
Similar princinple is applied in asking and recording tiie answers for tiie otiier
questions in part 5 and 6.
Total factor variance decisional balance items for confraception in general and lUD
Total Variance Explained
initial Eigenvalues
CotnpoasxA
Total
% of Variance
Cumulative %
Total
% of Variance
Cumulative %
5.6295
20.85
20.85
5.6295
20.85
20.85
3.536022
13.09638
33.94638
3.536022
13.09638
33.94638
2.153871
7.977299
41.92368
2.153871
7.977299
41.92368
1.712544
6.342757
48.26643
1.712544
6.342757
48.26643
1.38612
5.133777
53.40021
1.38612
5.133777
53.40021
1.099731
4.073076
57.47329
1.099731
4.073076
57.47329
0.973877
3.606953
61.08024
0.904202
3.348896
64.42914
0.880701
3.261854
67.69099
10
0.862067
3.19284
70.88383
224
Commune:
(Precontemplation group)
"We have used an lUD since we had our first baby. After 5 years, my wife had the
rUD removed to have our second child. After our second child, she had another lUD
inserted. I often think how lucky I am to have an lUD. I can enjoy my life more,
and do not have to worry about unwanted pregnancy."
WHAT ARE THE FACTS?
In our survey many of you knew of the different modem confraceptive methods
available in An Hai Distiict such as lUDs, tiie pill, condoms, and sterilisations, and
these can be easily obtained in the distiict health cenfre or commune health cenfres.
They are all effective methods. Traditional methods like withdrawal and periodic
abstinence are not very effective. About 30% of people using tiiese metiiods have
pregnancies that end in abortions.
Most of surveyed men know that lUDs help to prevent unwanted pregnancy. In fact,
lUD is one of the most effective revCTsible confraceptive metiiods in the world, with
225
effectiveness up to 99%, and minimum side effects. That means, only one person
can get pregnant among 100 women using lUDs. The lUD versions Tcu-380A
available in An Hai Distiict at this time can be effective for 10 years. lUD is very
convenient to use; your wife needs to have it inserted ONCE only, after which you
do not need to worry about unwanted pregnancy any more. It is provided free of
charge at any district health cenfre and commune health cenfre, and can be removed
any time upon request. An lUD can be inserted 42 days after a birth or an abortion,
it helps to prevent ectopic pregnancy when it is in place, and is very safe for women
with low risk of STDs or HFV/AIDS like women in the An Hong and Quoc Tuan
communes. NOW IT IS THE TIME FOR YOU TO THINK ABOUT lUDS.
CAN IT HAPPEN TO YOU?
About one-fourth of men in An Hong and Quoc Tuan communes have wives that
have undergone abortion; and,for a similar number of you, your last births were
unwanted. Many of you did not use effective confraceptive methods at all, because
you you have sex irregularly or because you think you are too old. As you know,
pregnancy can occur in 85% of all couples who are not using contraception, and
many unwanted pregnancies end in abortion. According to worldwide statistics,
about 350 women die per 100,000 induced abortions. About one-third of
Vietnamese women who have had abortions have health problems, such as
reproductive fract infection, secondary infertility, and miscarriage. What happens to
poor unwanted children? They often lack parental care because parents have neitiier
time nor resources to take care of them. Will those children have good educational
opportunities? Moretiianlikely,tiieywill not.
"I WANT MY WIFE TO TRY AN lUD TO SET A GOOD EXAMPLE FOR MY
FAMILY AND MY FRIENDS"
You would do anytiiing to keep your family happy and healtiiy, because you are a
pillar of tiie family. Do you know tiiat two-tiiirds of married men in Quoc Tuan and
An Hong communes have wives who use lUDs. You are not so different from most
of the other men in tiiese communes. You probably know a lot about lUDs and
other confraceptive methods. Maybe you think you will be embarrassed if you talk
about confraception with your wife or otiier people. Don't be. ft may help if you
226
tiiink of confraception as just anotiier topic of daily discussion. And think of how
much it is going to help you and your wife - no more worries about unwanted
pregnancies. That is important, because your efforts to keep your family happier and
healthier can make a difference. By sharing this, you are making a good model
husband.
You are not thinking about using an lUD now, but that may change in the future. If
it does, you will find an lUD will help make your life more enjoyable, and help
ensure a good future for your children. That's why the government of Vietaam
encourages couples to use lUDs as part of its family planning program.
Please contact us if you have any more enquiries.
HANOI SCHOOL OF PUBLIC HEALTH.
138 GIANG V, BADINH, HANOI.
Tel/Fax: 84 4 8231 743
227
Commune:
(Contemplation/Preparation group)
" We have had lUD since we have first baby. After 5 years, my wife took out lUD
to have my second child. After second child, she has lUD inserted again. I have
thought I am so lucky to have an lUD. I can enjoy my life more, and do not have to
worry about imwanted pregnancy"
WHAT ARE THE FACTS
In the survey you filled out, you told us you would tiiink about convincing you wife
to adopt an lUD. We think that's great. Did you know in our communes, about onefourth of men have wives who have undergone abortion and whose last birth was
unwanted. Many of you did not use effective confraceptive methods at all, because
you were having sex irregularly or because you thought you were too old. As you
know, pregnancy can occur in 85% of all couples who are not using confraception,
and many unwanted pregnancies end in abortion. About 350 women die per
100,000 induced abortions. About one-thfrd of Vietnamese women who have had
abortions have health problems such as reproductive fract infection, secondary
infertility, and miscarriage. What happens to poor unwanted children?. They often
lack care, because parents have neither time nor resources to take care of them. Will
228
tiiose children have good educational opportunity in tiie future? ft is unlikely tiiat
they will.
In our survey many of you knew about fraditional metiiods like withdrawal and
periodic abstinence in addition to modem confraceptive metiiods. But fraditional
methods are often not very effective. About 30% of people using these methods
have pregnancies that end in abortions.
WHY AN lUD? WEIGH THE BENEFITS
You may have heard that lUDs can cause bleeding and pain. While it is true tiiat
after insertion, in the first few months, women can have abdominal pain, irregular
menstrual periods, but these symptoms will disappear in tiie next two or three
months, and tiiey are not signs of illness. DON'T LET THESE RUMOURS HOLD
YOUBACK!
"I WANT MY WIFE TO TRY AN lUD TO SET A GOOD EXAMPLE FOR MY
FAMILY AND MY FRIENDS"
You would do anything to keep your family happy and healthy, because you are a
pillar of the family. Do you know that two-thirds of married men in Quoc Tuan and
An Hong communes have wives who use lUDs. You are not so different from most
of other men in these communes. You probably know a lot about confraception and
lUDs. Maybe you think you will be embarrassed if you talk about lUDS with your
wife or other people. Don't be. It may help if you think of using lUDs as just
another topic of daily discussion. And think of how much it is going to help you and
your wife - no more worries about imwanted pregnancies. That is important,
because your efforts to keep your family happier and healthier can make a
difference. By sharing your knowledge and opinions, you are making yourself a
good model husband.
Please contact us if you have any more enquiries.
HANOI SCHOOL OF PUBLIC HEALTH.
138 GL\NG VO - BADINH -HANOI.
Tel/Fax: 84 4 8231 743
229
.Commune:
(Action/Maintenance group)
Please see print copy for image
NGHIEN C(J[a
VAI TRO C O A NAM GI61 TRONG CONG TAC SvHHCD TAI HUYEN AN HAI
" ' ' K M i g u i anh:
Xa:
Thon:
r!,?C^'S:'^^^-"^>'5i^=^"^At^s=3ijSS:3??t--Si;'r-ri=??i-"*- "-w^
MOT SO VAN Bi LifiN QUAN DfiN CAC BIEN PHAP TRANH THAI (BPTT)
Trong nghicn cihi cua chiing toi, h^u het cac doi tuang phong v& din biet nhjgu ve
bien phap iranh thai hien dai dang dugc cung ca'p a huyen An Hai nhu vong tranh thai,
thuoc none tranh thai, bao cao su va triet san. Day la nhOtig bien phap tranh thai hien dai
CO hieu qua \a luon c6 sSn tai trung tarn y te huyen hoac tram y te xa. Ben canh do, nhfing bien phap
tranh thai iru\t^n thong nhu xuat tinh ngoai am dao hay tinh vong kinh d6u c6 hieu qua
tuofng doi ihap. Co tdi 30% &6 nguod ap dung bien phap tranh thai tmy6n thong da c6 thai
ngoai y inuon \a ho da phai di nao pha thai..
Hau liei nhfmg ngmri dan ong khi dugfc phong van chi biet rSng vong tranh thai giiip
ngan ngCra vice c() thai ngoai y muon. Tren thuc te, vong tranh thai la mot bien phap c6
hieu qua tninli thai vao loai cao nhat tren th6' gi6i, hitu qua ciia no dat ted 99% va it c6
tac dung phii. Dit^u do c6 nghia la cii trong 100 nguod sir dung vong tranh thai,
chi CO 1 nguo'i c6 the' c6 thai ngoai y muon. Loai vong tranh thai TCU-380A - hien dang
luu hanh (V huycn;4An Hai - c6 hieu qua tac dung trong vong 10 nam.
Vong tninh thai rat thuan tien trong viec su dung, ngudi phu nu chi can dat vong mot
Ian nliung no c6 tac dung phong tranh thai trong nhi6u nam. Nhu vay, cac cap vg chOng
thong phiii Io lang vc viec c6 thai ngoai y muon. Vong tranh thai dugc cung cap rong rai
tai tat ca cac trung tani y te va cac tram y te xa cua huyen An Hai. Vong tranh thai c6 th6
thao ra bat cii liic nai:> neu muon. Hon niJa, vong tranh thai c6 the dat ngay sau khi sinh con
42 ngay hoac sau nao hut thai cho nen rat c6 Igi trong viec phong tranh thai.
Vong tranh thai la mot bien phap rat an toan cho nhirng phu nif khong bi mSc cac benh
lay t n n e n qua dudiig tmh due hoac HIV/AIDS, gid'ng nhu nhiing nguai phu nil tai
hai xa An H6ng va Quoc Tuan, huyen An Hai, T P H a i Phong
MANG THAI NGOAI Y MUON CO THE XAY RA VCfl VO CHONG ANH KHONG
"101 Ml ON V 0 MINH DAT \ ON(i TRANH TH VI DE L4M GI ONG (HO (ilA DINH VA BAN BE."
Nguai dan ong thucmg lam moi viec de gia dinh dugc hanh phuc va khoe manh vi ho la
nhfrng thanh vien tru c6t ciia gia dinh. Anh biet khong, c6 toi 2/3 s6 cap vg ch6ng o xa
An Hong va xa Quoc Tu&i da ap dung cac bien phap tranh thai de ke hoach hoa gia dinh.
Vg ch6ng anh ciing c6 th6' diing vong tranh thai vi anh chi cung giong nhu cac cap
vg ch6ng khac trong xa. Anh nen nghi den vong tranh thai di ke hoach hoa gia dinh.
Mai dau co the anh se cam thSfy xSiu ho khi thao luan voi vg hoac nhiing nguai xung quanh
ve vong tranh thai, nhung sau mot vai Ian chiing toi tin ring anh se quen ngay. Anh dimg
ngai ngimg khi thao luan v6 vSii de nay. Thao luan voi vg va nhung ngudi xung quanh ve
\6ng iriinh thai la rat hiru ich neu nhu anh cho rang no cung gidng nhu thao luan ve cac
\an de khac trong cugc song hang ngay. Viec trao doi va thao luan voi vg va nhiing
ngudi khac \6 vong tranh thai la rai can thiet vi no cd the lam cho cugc song cua gia dinh
anh hanh phiic va tot dep han. Khi anh chia xe thdng tm v^ bien phap tranh thai vdi vg thi
anh da thirc sir la mot ngudi chong hien dai cua thien nien ky mdi.
Hic'n uii CO [he UIL/I chua nghi den viec su dung vdng tranh thai nhung sau nay thi rat
CO the. Neil fihif vo chong anh dp dung vdng tranh thai, anh se thay vong tranh thai
giup cho CHOC song cila anh chi hanh phuc hem vd dam bdo mot tuang lai tot dep cho
con cdi anh chi. Chinh vi vay, chinh phu Viet nam ddkhuyen khich cac cap vg chong su dung
vong tranh thai trom; chUimg trinh dan so-ke hoach hod gia dinh.
Neu cd thac mac gi, hoac chua rd ve vdng tranh thai, hay lien lac vdi chung tdi theo dia chi:
NQHiEN ccjra
VAI TRO CUA NAM Cidl TRONG CONG TAC KHHGO TAI HUYEN AN HAI
Kinh giii anh:
Xa:
/^'
Thon:
^^
MOT SO VAN DE LIEN QUAN DEN CAC BIEN PHAP TRANH THAI (BPTT)
ir i~^.. t ^ ./ ^ i _ w a o a ^
<.'::;.r<.aife.
&-^^:^^^^S'S^''^~_
Khi chiing tdi phong van anh, anh cd ndi vdi chiing tdi la se thuyei phuc chi nha di dat
vong tranh tliai. Chiing tdi tha'y dieu dd that tuyet vdi. Anh biet khdng, khoang 1/4 cac ba vg
o- hai \ a .An Hong va Qudc Tuan da tung di nao pha thai hoac sinh diia con lit la do bi vd
kc hoach. Rai nhidu ngudi da khdng sir dung cac BPTT hien dai bdi ho nghi tuoi ho da cao
lia\ hieni khi sinh boat tinh due nen khdng the cd thai dugc.
Nhimg thuc to, cd IcVi 85% so cac cap vg chong khdng sir dting BPTT la cd nguy co cd thai
\a hang loat cac trudng hgp cd thai ngoai mong mudn deu di den giai phap la nao hiit thai.
Cii trong 100.000 ca nao hiit thai thi cd tdi 350 ca bi tir vong. C6 khoang 1/3 so phu nii
nao hiit thai c^ Viei nam da gap phai nhung van de siic khoe nhu: nhiSm khuin dudng
sinh due, dm yen va c6 the din den vd sinh.... Va dieu gi se xay ra doi vdi nhiing diia tre
sinh ra do bi Id ke' hoach.' Chiing thudng thieu vang su cham sdc cua cha me vi cac bac
cha mc thudng khdng eo nhieu thdi gian hoac cd dieu kien kinh te de cham sdc con cai.
Nhu vay. lieu rfing nhung diia tre nay se cd co hdi dugc giao due tot trong tuang lai khdng?
Cau tni Icti d day thudng la "Khdng".
Qua nghien euu. rat nhicu nam gidi biet rang ben canh cac BPTT hien dai con cd BPTT
tniycn thdng nhir xuat tinh ngoai am dao hay tinh vdng kinh. Nhung cac BPTT truyen thdng
cd hieu qua tucnig ddi ihap trong viec tranh thai. Cd tdi 30% sd ngudi ap dung bien phap
tranh thai truvdn tltdns da co thai ngoai y mudn va ho da phai di nao pha thai.
._
_ NH^^
"TOI MUON VO MINH DAT VONG TRANH THAI DE LAM GlTONG CHO GIA DINH VA BAN BE."
Ngudi dan dng thudng lam moi viec de gia dinh dugc hanh phiic va khoe manh vi ho la
nhiing thanh vien tru cot cua gia dinh. Anh biet khdng, cd tdi 2/3 sd cap vg chong d xa
An Hong va xa Qudc Tuan da ap dung cac bien phap tranh thai de ke hoach hoa gia dinh.
Vg chong anh ciing cd the dung vdng tranh thai vi anh chi ciing gidng nhu cac cap vg chong
khac trong xa. Anh nen nghi den vdng tranh thai de ke hoach hoa gia dinh. Mdi dau cd the
anh se cam thay xau ho khi thao luan vdi vg hoac nhirng ngudi xung quanh v6 vdng tranh
thai, nhung sau mot vai Ian chiing tdi tin ring anh se quen ngay. Anh dtmg ngai ngiing khi
thao luan \i van di nay. Thao luan vdi vg va nhiing ngudi xung quanh ve vdng tranh thai la
rai hiiu ich neu nhu anh cho ring no ciing gidng nhu thao luan ve cac van de khac trong
cugc song hang ngay. Viec trao ddi va thao luan vdi vg va nhirng ngudi khac ve vdng.
tninh thai la rat can thiet vi no cd the lam cho cugc sdng cua gia dinh anh hanh phiic va
tot dep hctu. Khi anh chia xe thdng tin ve bien phap thai vdi vg thi anh da thuc su la mot
ngiTcfi chong hien dai cua thien nien ky mdi.
Neil cd thac mac gi. hoac chua ro ve vdng tranh thai, hay Hen lac vdi chiing tdi theo dia chi:
TRLCJNG DAI HOC Y TE CONG C O N G
^ ^ 138 Giang V5- Ba Dinh - Ha Noi
BiiiThiThuHa
Tel: 04-8231743
^s^eiliSa-'SSs:&2^fesS4^&a>
NGHIEN Cda
VAI TRO CUA NAM GI61 TRONG CONG TAC KHHGO TAI HUYEN AN HAI
Kinh giri anh:.
Xa:
Thon:
H-f-.
abfe^liaS^^:^'.J2i-JiJvCtfialr.:^i^Sg^
fcMo^iimg anh da chon vong tranh thai de ke hoach hoa gia dinh.
^ ^ a n h chi co cam tha'y an toan khi diing vong tranh thai khdng?
Chiing tdi hy vong ring:
vdi sir lua chgn ciia minh va se tiep tuc duy tri sir dung vdng tranh thai,
iing kinh nghiem cua minh ve sir dung vdng tranh thai vdi moi ngudi.
oac chua rd ve vdng tranh thai, hay lien lac vdi chiing tdi theo dia chi:
231
TRAINING PROGRAM
MEN'S READINESS TO ACCEPT lUD FOR CONTRACEPTION
IN RURAL VIETNAM
Time
11-13/11/2001 (first round)
9-11/3/2002 (second round)
Hanoi 2001
CONTENTS
1. Brief about the research project
1.1. Objectives
Overall objectivs
To promote men's participation in reproductive health and the family planning
decision-making process
Specific objectivse
1. To increase men's awareness on contraceptive methods, particularly modem
methods
2. To increase lUD use
3. To reduce myths/perception related to side effects of lUDs
4. To reinforce couple communication and communication with other people in the
commimity network on family planning and lUDs
5. To create opportunities for people to facilitate the behavior changes in lUD
acceptance among men and couples
1.2. Plan for intervention
Three villages in each commune will be chosen for the intervention and three villages
in each commune will be selected for the control group. A total of 6 villages will be
in the intervention group and 6 villages will be in control group. The intervention
group will receive appropriate messages corresponding to the stages of men's
readiness to accept lUD for contraception. The control group will not receive any
message. The evaluation will be carried out after 6 months
Intervention group: total number of men in each village and stage
Commune Intervention
Stage 1
Stage 3
Stage 2
village
VanXa
Quoc
12
29
9
Tuan
Kieu H a l
9
2
2
(X7)
27
Kieu Ha 2
8
6
(X6)
48
Tat Xung
An Hong
32
10
42
2
Ngo Himg
16
61
8
Le Sang
31
216
39
99
Taeng
Total
50
13
41
90
60
100
354
Control villages
Quoc Tuan: Nhu Kieu 1, Nhu Kieu 2, Kieu Thuong
An Hong: Pham Dung, Le Lac 2, Le Lac 1
The design of the messages was developed from the results of baseline survey in April
2001. Three different messages will be provided to men in 3 different stages of
readiness: precontemplation (stage 1); contemplation/preparation (stage 2); and
action/maintenance stage (stage 3). Each of the messages is relevant to the men's
stage of change.
1.3. Intervention plan
Two intervention rounds will be carried out in 6 months.
The first round will be carried out in November 2001. Objective of this round is to
motivate men to participate in the study, to reduce misconceptions about side-effects
and increase awareness of the effectiveness of lUD, to help them to start looking at
the lUD option.
The second module will be implemented after 3 months (about in February 2002). All
participants will receive relevant information for their stage of change for lUD.
Objective of this module was to reinforce the change, promote communication and
acceptance of lUD
2. Intervention activities
November: first round
First home visit
First return home visit
February: second round
Second home visit
Second return home visit
The content of each visit is decribed in the intervention protocol.
3. Supervision
Supervisor should follow the program of previous data collection phases. A total of 20
percent of participants will be randomly checked to see whether interviewer was
correctly performing the intervention protocol. In cases where 5 percent of
participants of one interviewer was not correctly administrating the procedure, that
interviewer will be requested to do interview again all participants that he/she is in
charge of
4. Training program on intervention activities
4.1. Objectives
After this training program, the collaborators should be able
1. To imderstand the procedures offirstintervention round (2 home visits in
November).
2. To understand the procedures of second intervention round (2 home visits in
March).
3. To implement all activities required for the intervention round
4. To conduct all activities under supervision of district health workers.
4.2. Contents
4.2.1. To introduce the intervention program: objectives, time, select study
population and study group.
4.2.2. To explain how to identify people in different groups by using staging
algorithm
Stage 1 (precontemplation): people are not ready to change, they lack basic
knowledge of contraceptive metiiod, do not know abouttiieprobability of
getting pregnancy and severity of risks of abortion
Stage 2 (contemplation/preparation): people intend to change (in 30 days to 6
months)
Stage 3 (action/maintenance stage): people have already changed behavior and
have been using lUD for at least 6 montiis.
4.2.3. Intervention components
- Stage-targeted letters. The objective of letters was to provide information to
men on lUD method. The content of letters was developed following the
results of baseline in April 2001.
-
Sreening form (1.1): to check the stage of change for lUD use
Consent form: (form 1.2): consentee is willing to participate in the study, and
knows he can withdraw at any time ifhe does not want to continue. Those
who accept to participate in the study need to sign the form
Notification of home visit (form 1.3): the form should record the time (date)
that interviewer visits the men, the proposed time for the next visit and
supervision comments.
Questionnaire to assess men's perception regarding stage-targeted letters
(form 1.4).
INTERVENTION PROTOCOL
Overview
1 Prepare list of all members in intervention group with respective stages
2 Assign family planning collaborators (FPC) to villagetiiattiieywill be in charge
of
3. Train FPC in distiict health centi-e (DHC) on counselling and procedure for each
intervention visit
4. Prepare field supervision guideline
Provide incentives for participants after first and second visit
Complete intervention module 1: first visit and return visit
Complete the evaluation form of targeted letters 1
Complete intervaition module 2: first visit and retum visit
Complete the evaluation form of targeted letters 2
Date
October: module 1
First visit
First retum visit after one week
January: module 2
First visit
Second retum visit after one week
List of FPC in charge of each village
Village
Name of FPC
1. VanXa-Dl
1.
2. Kieu Ha-X6
3. Kieu Thuong
4. Tat Xung
5. Ngo Hung
6. LeSang
Consent form
Ifthe participant agreed to participate in this trial, the intervention must be
performed.
At the end of the first or second visit, arrange time with participant for retum
visits
Monitoring form
- To record any event in the local area related to family planning activities (lUD
campaign, any contest related to family planning in the village, change of local FP
collaborators etc) that could affect the results of intervention.
Precontemplation group
1- Greeting
2- Ask for the questions
3- Try to answers questions
> Question regarding side effects (the most concem) -> tiy to convince men
tiiese are temporary problems and will go away in time.
> Compare the costs of bearing some minor side effects with other costs such as
worries of getting unwanted pregnancy, or abortion. List the advantages of
lUD, e.g. time saved for other works etc.
> Compare the advantages of lUD with otiier modem methods available in the
district (convenience, costs, side effects, sexual pleasure, social acceptance)
> Ask men to think about lUD option
4- For the question that can not be answered, please record and promise to give them
feedback very soon.
> There is a possibility to get the feedback by print materials or in person
contact (person in charge of FP program will come and discuss with them)
5- Complete the evaluation questionnaires of the tailoring message
II. Contemplation / Preparation group
1- Greeting
2- Ask for the questions
3- Try to answers questions
> Question regarding side effects (the most concem) -> try to convince men
these are temporary problems and will go away in time.
> Compare the costs of bearing some minor side effects with otiier costs such as
worries of getting unwanted pregnancy, or abortion. List the advantages of
lUD, e.g. time saved for other works etc.
> Compare the advantages of lUD witii other modem methods available in the
district (convenience, costs, side effects, sexual pleasure, social acceptance)
> Convince men to undertake a trial and to check with other people who have
usedlUD
10
4- For the question that can not be answered, please record and promise to give them
feedback very soon.
> There is a possibility to get the feedback by print materials or in person
contact (person in charge of FP program will come and discuss with them)
5- Complete evaluation questionnaires of the tailoring message
III.
11
CR
AM
Intervention module 2
(after 3 months intervention)
P-^ CR (positive group)
P-^ AM (positive group)
P^P
CR-> AM (positive group)
CR^CR
CR-^ P (relapse group)
A M ^ AM
AM-> CR (relapse group)
AM-> P (relapse group)
12
Arrange group discussion for those people, and speaker will be successful adopter
oflUD.
Try to convince the advantages of lUD over other methods, and costs of lUD use
with other costs (unwanted pregnancy, abortion)
Try to compare the costs of lUD use with other costs (unwanted pregnancy,
severity of abortion etc)
Try to convince men with advantages of lUD over other methods (traditional not
effective, other methods: inconvenience, etc)
Ask for experiences of getting lUD, and questions related to lUD use.
13
14
15
4. Convince men to accept a trial and to check with other people that have used lUD
5. Complete evaluation questionnaires ofthe tailoring message
III. Action/ Maintenance group
1. Rewarding of adoption of lUD
2. Try to answer the questions
3. Those questions that can not be answered, try to record and will provide feedback
soon
4. Complete evaluation questionnaires ofthe tailoring message
16
My name is
. I am willing to take part in this
research study. I amfiiUyaware that I can drop the study at any time I want to.
Signature
17
18
An Hong:
Date of of date :
First visit
Second visit
Date:
Date:
Date:
Date:
Time:
Time:
Time:
Time:
Results:
Results:
Results:
Results:
Complete
Complete
Complete
Complete
Absent
Absent
Absent
Absent
Postpone
Postpone
Postpone
Postpone
Refuse
Refuse
Refuse
Refuse
Other
Other
Other
Other
Supervision
Supervision
Supervision
Supervision
Date
Date
Date
Date
Results
Results
Results
Results
Good
Good
Good
Good
Acceptable
Acceptable
Acceptable
Acceptable
Poor
Poor
Poor
Poor
Requestfor
Requestfor
Requestfor
Requestfor
re-administrate
re-administrate
re-administrate
re-administrate
Spot check
Spot check
Spot check
Spot check
Name
Name
Name
Name
Date:
Date:
Date:
Date:
Results
Results
Results
Results
Additional notes
19
2- No
1 - Yes
2- No
1- Yes
2- No
2-No
Describe in details::
1-Yes
1 - Yes
2-
2- No
Q1.2
Q1.3
2
Q2.1
Q2.2
3
Q3.1
Q3.2
Q3.3
Q3.4
Q3.5
Q3.6
Q3.7
4
Q4.1
Q4.2
Q4.3
5
Q5.1
Q5.2
Q5.3
Q5.4
6
Q6.1.
Q6.2
Q6.3
233
Percentage (%)
(N=327)
Effectiveness of lUD
100.0
99.7
Pregnancy susceptibiUty
93.0
Abortion severity
98.8
Communication on lUD
69.4
Adoption of lUD
63.0
Otiiers
26.0
Mean
SD
6.1
0.7
6.1
0.7
6.1
0.7
234
Understanding \e\^\
8 60.0
o>
3 50.0
g 40.0
Low
D Moderate
DHigh
I 30.0 20.0
10.0
0.0
RELEVAN1
3
O
.Q
CO
SB
Friends/Relati\^s
Health workers
Other peopie
sWix^s
.<. %
Q -
SI
CO
Percentages
235
iSucceded in con^ncing
I Total men in each group
To get IUD
To continue use
IUD
236
at baseline
study
Characteristics
Age group
19-24
25-29
30-34
35-39
40-44
45-49
Education
Primary
Lower secondary
Upper secondary
Higher education
Occupation
Government
Private
Agricultural
Household
Other
Parity
One
Two
Three and more
Having a son
Had no son
Had son
Desired children
Not desire more
Desire more
Having abortion
No
Had abortion
Last birth wanted
No
Yes
Evaluation group
Lost to follow -up Included
(N=41)
(N=610)
Total
0
31.7
29.3
24.4
14.6
2.3
14.4
22.8
35.2
23.0
2.3
2.2
15.5
23.2
34.6
24.6
7.3
53.7
34.1
4.9
7.6
61.0
25.1
6.4
7.6
60.5
25.7
6.3
0.63
14.6
19.5
63.4
2.4
16.4
16.9
61.6
2.3
2.8
16.3
17.1
61.8
2.3
2.6
0.8
68.3
24.4
7.3
28.5
53.1
18.4
31.0
51.3
17.7
0.00**
39.0
61.0
25.1
74.9
26.0
74.0
0.04**
58.5
41.5
27.2
72.8
29.2
70.8
0.00***
24.4
75.6
25.9
74.1
25.8
74.2
0.5
19.5
80.5
27.4
72.6
26.9
73.1
0.2
P values
na
237
Communication with
others
Low
High
Communication with
wives
Low
High
Spontaneous recall
Modem metiiod
Traditional method
SOC for IUD use
Precontemplation
Contemplation/
Preparation
Action/Maintenance
Study group
Intervention
Control
97.6
2.4
95.7
4.3
95.9
4.1
0.5
65.9
34.1
78.7
21.3
77.9
22.1
0.04**
95.1
41.5
93.9
38.2
94.0
38.4
0.7
0.3
39.0
28.9
29.5
0.3
7.3
53.7
10.8
60.3
10.6
59.9
43.9
56.1
55.1
44.9
54.4
45.6
0.11
238
239
240
1.
Commune:
2.
Village:
3.
4.
5.
Household
number:
Name of men
Group
6.
ID number
An Hong (2)
1 Kieu ha-X7
4 NhuKieu
1 Pham Dung
4 LeLac 1
2Kieuha-X6
5 Nhu Kieu
2 Tat xung
5 Ngo Hung
3 Vanxa-Dl
6 Kieu thuong
3 LeLac2
6 Lesang
Intervention 1: 1.3 (CI); 1.4 (C2); 1.6 (C3); 2.2 (C4); 2.5 (C5); 2.6 (C6)
Control 2: 1.1 (Dl); 1.2 (D2); 1.5(D3); 2.1 (D4); 2.3 0)5); 2.4 (D6)
C V G MONTH YEAR BIRTH
Interview result
Interview
Day/month/year
/
Completed
Respondent absent
Postponed
Refused
Other (specify)
5.
Remarks
Spot checked by
Name
Date
YES 1
NO 2
YES 1
NO 2
YES 1
NO 2
YES 1
NO 2
YES 1
NO 2
YES 1
NO 2
NO
QlOl
SKIP TO
/_]
98
99
DK
NO RESPONSE
YEARL_/_J
DK
98
NO RESPONSE
99
Q102
Q103
Q104
Q105
Q106
Q107
Q108
KINH 1
HOA 2
MUONG 3
NUNG 4
HMONG 5
OTHER (SPECIFY)
MIXED ETHNICITY 8
NO RESPONSE
99
Q109
QUO
QUI
Q112
[_/_|
98
99
DK
NORESPONSE
98
99
TOTAL BIRTHS:
NON
DK
NO RESPONSE
0
98
99
]=Q106
Q113
Q114
Q115
Q116
Q117
Q118
TOTAL SONS:
TOTAL DAUGHTER
TOTAL CHILDREN
NON
0
DK
98
NO RESPONSE
99
YEARS:
MONTHS:
DK
98
NO RESPONSE
99
YES
I
NO
2
DK
98
NORESPONSE
99
TOTAL ABORTION:
NON
0
DK
98
NO RESPONSE
99
AGE IN COMPLETED YEARS [_/_]
DK
98
NO RESPONSE
99
BUDDIES 1
CATHOLIC 2
CHRISTIAN 3
OTHER (SPECIFY)
NO RELIGION 0
NO RESPONSE
99
Q119
KINH 1
HOA 2
MUONG 3
NUNG 4
HMONG 5
OTHER (SPECIFY)
MIXED ETHNICITY 8
NO RESPONSE
99
Q120
GOVERNMENT STAFF
1
PRIVATE BUSINESS 2
AGRICULTURAL/FARMING JOB 3
HOUSE-HUSBAND 4
OTHER (SPECIFY)
DK
NO RESPONSE
Q121
Q201
98
99
PRIMARY
1
LOWER SECONDARY 2
UPPER SECONDARY 3
HIGHER
4
DK
98
NO RESPONSE
99
CIRCLE ONE
SISCTION 2
FAMILY PLAIWEVG KNOWLEDGE
SEVERAL DAYS BEFORE MENSTRUATION
Which days ofthe woman's
PERIOD
1
menstrual period do you feel are
DURING MENSTRUATION PERIOD 2
SOME DAYS AFTER
safe to have intercourse with your
MENSTRUATION PERIOD
3
wife if you do not want her to
MID-CYCLE BETWEEN MENSTRUATION
conceive?
PERIOD
4
OTHER rSPECIFY)
DK
NO RESPONSE
98
99
^Q117
Q203
YES 1
NO 2
-> Section 3
YES 1
NO 2
-> Sections
Q204
PILLS I
CONDOMS 2
IUD 3
VASECTOMY 4
TUBECTOMY 5
INJECTABLES 6
RHYTHMS 7
WrrHDRAWAL8
OTHERS (SPECIFY)
DK
98
NO RESPONSE
99
Q205
PILLS 1
CONDOMS 2
IUD 3
VASECTOMY 4
TUBECTOMY 5
INJECTABLES 6
RHYTHM 7
WITHDRAWAL 8
OTHERS (SPECIFY)
DK
98
NO RESPONSE
99
Q206
Q207
RADIO
I
TELEVISION
2
NEWSPAPERS
3
BILLBOARDS
4
HEALTH WORKERS
5
FAMILY PLANNING COLLABORATOR 6
FRIENDS /RELATIVES 7
WOMEN'S UNIONS 8
WIFE9
OTHER (SPECIFY)
DK
98
NO RESPONSE
99
SIDE EFFECT 1
EFFECTIVENESS
2
HOW TO USE
3
INDICATIONS
4
OTHER (SPECIFY)
DK
NO RESPONSE
98
99
SECTION 3
COMMUNICATION ON FAMILY PLANNING
Q301
NEVER
0
Yes
1-3 times
> 3 times
DK
NO RESPONSE
98
99
NEVER
0
Yes
1-3 times
> 3 times
DK
NO RESPONSE
Q303
NEVER
0
Yes
1-3 times
> 3 times
DK
NO RESPONSE
Q304
Q306
0
Yes
1-3 times
> 3 times
Q308
NEVER
0
Yes
1-3 times
> 3 times
0
Yes
1-3 times
> 3 times
DK
NO RESPONSE
98
99
NEVER
0
Yes
1-3 times
> 3 times
98
99
NEVER
0
Yes
1-3 times
> 3 times
DK
NO RESPONSE
Q310
98
99
NEVER
DK
NO RESPONSE
Q309
98
99
NEVER 0
TIMES
DK
98
NO RESPONSE
99
DK
NO RESPONSE
Q307
98
99
NEVER
DK
NO RESPONSE
Q305
98
99
98
99
NEVER
0
Yes
1-3 times
> 3 times
DK
NO RESPONSE
98
99
Q401
SECTION 4
FAMILY PLANNING PRACTTrE
Are you currently using any method
to delay or avoid having a child?
Q402
Q403
YES 1
NO 2
PILLS 1
CONDOMS 2
IUD 3
VASECTOMY 4
TUBECTOMY 5
INJECTABLES 6
RHYTHMS 7
WTTHDRAWAL8
OTHERS (SPECIFY)
DK
98
NO RESPONSE
99
TO HAVE ANOTHER CHILD SOONl
DIFFICULT TO GET PREGNANT 2
I AM TOO OLD 3
STERILE 4
TOO MUCH TROUBLES 5
FEAR OF SIDE EFFECTS 6
TOO EXPENSIVE 7
RELIGIOUS OBJECTION 8
IRREGULARTTY 9
UNAVAILABTTLFTY 10
OTHER (SPECIFY)
DK
NO RESPONSE
Q404
Q405
Other
HAVE A SON
HAVE A DAUGHTER
NO DIFFERENCE
HAVE BOTH SONS AND DAUGHTER
OTHER (SPECIFY)
Q409
98
99
> 2 YEARS 1
=<2YEAR
2
CAN'T GET PREGNANT 3
OTHER SPECIFY
DK
NO RESPONSE
Q408
98
99
Q407
98
99
Total number
DK
NO RESPONSE
Q406
->Q402
"Q403
98
99
THEN 1
LATER 2
NOT AT ALL 3
< 6 months
=> 6 months
AC 1
M 2
^Q408
Q501
Q502
Q503
Q504
Q506
Q507
Q509
Q510
Q512
Q513
Not
important
2
Very
important
Important
Do not
know
important
2
know
3
important
5
Q511
important
4
Q508
know
3
Benefits of contraception
How important is each of these advantages to you in deciding whether or not to use family
planning methods for prevention of pregnancy using 5 point scale
FP helps you to be more
responsible for decisions about
having children
FP helps you avoid the results
of unwanted pregnancy
Your wife would not have to
worry about becoming pregnant
if you were using contraception
FP helps to limit size ofthe
family
How important is each of these advantages to you in deciding whether or not to use
family planning methods for prevention of pre gnancy using 5 point scale
Costs of contraception
Not very
important
1
Q505
important
2
Q514
Confident
Do not
know
Coafidait
Q601
Q602
Q603
Q604
Q605
Q605a
Yes=l
No = 2
2
Q606
Q607
Q608
Q609
Veiy
confident
*Q610
-Q606
Very
confident
Not
confident
1
2
Do not
know
confident
Very
confident
Q610
Q611
Q612
Q613
2. The letter from the Joumal 'Health Education Research' on paper's acceptance
241
HEALTH
EDUCATION
RESEARCH
THEORY & PRACTICE
Published by Oxford University Press
Great Clarendon Street, Oxford 0X2 6DP, UK
Tel 01865 556767 Fax 01865 267773
DrBmThiThuHa
Hgjxoi School of Public Heafeh.
138 Giang vo
Badinh
Haiioi
Vietnam
, &
Ecutlve Editor
'^f. K. Tones
}2 Moseley Wood Lane
ookridge
dsLS16 7EP
10113 2674706
10113 2619526
'ii Keith.tones@virgin.nel:
Associate Editor
J. K. Davies
University of Brighton
School of Applied Social Science
Falmer
Brighton BN1 9PH
Tel 01273 643476
Fax 01273 643473
Email j.k.davies@bton.ac.ulc
Dr Alan S. Rigby
Dept. of Paediatrics
University of Sheffield
Sheffield SI 0 2TH
Tel 0114 2717120
Fax 0114 2755364
Email a.s.rigby@sheffield.ac.uk